id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 3
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-12-13 00:00:00
| authors_parsed
listlengths 1
3.44k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2504.16163
|
Himansh Rathore
|
Himansh Rathore (U. Arizona), Gurtina Besla (U. Arizona), Kathryne J. Daniel (U. Arizona), Leandro Beraldo e Silva (Obs. Nacional)
|
Response of the LMC's Bar to a Recent SMC Collision and Implications for the SMC's Dark Matter Profile
|
26 pages, 15 figures, accepted to ApJ, updated to the accepted version
| null |
10.3847/1538-4357/ade0ae
| null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
The LMC's stellar bar is offset from the outer disk center, tilted from the disk plane, and does not drive gas inflows. These properties are atypical of bars in gas-rich galaxies, yet the LMC bar's strength and radius are similar to typical barred galaxies. Using N-body hydrodynamic simulations, we show that the LMC's unusual bar is explainable if there was a recent collision (impact parameter $\approx$2 kpc) between the LMC and SMC. Pre-collision, the simulated bar is centered and co-planar. Post-collision, the simulated bar is offset ($\approx$1.5 kpc) and tilted ($\approx8.6^\circ$). The simulated bar offset reduces with time, and comparing with the observed offset ($\approx0.8$ kpc) suggests the timing of the true collision to be 150-200 Myr ago. 150 Myr post-collision, the LMC's bar is centered with its dark matter halo, whereas the outer disk center is separated from the dark matter center by $\approx1$ kpc. The SMC collision produces a tilted-ring structure for the simulated LMC, consistent with observations. Post-collision, the simulated LMC bar's pattern speed decreases by a factor of two. We also provide a generalizable framework to quantitatively compare the LMC's central gas distribution in different LMC-SMC interaction scenarios. We demonstrate that the SMC's torques on the LMC's bar during the collision are sufficient to explain the observed bar tilt, provided the SMC's total mass within 2 kpc was $(0.8-2.4) \times 10^9$ M$_\odot$. Therefore, the LMC bar's tilt constrains the SMC's pre-collision dark matter profile, and requires the SMC to be a dark matter-dominated galaxy.
|
[
{
"version": "v1",
"created": "Tue, 22 Apr 2025 18:00:01 GMT"
},
{
"version": "v2",
"created": "Sat, 24 May 2025 04:42:07 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Rathore",
"Himansh",
"",
"U. Arizona"
],
[
"Besla",
"Gurtina",
"",
"U. Arizona"
],
[
"Daniel",
"Kathryne J.",
"",
"U. Arizona"
],
[
"Silva",
"Leandro Beraldo e",
"",
"Obs. Nacional"
]
] |
2504.16558
|
Juan Carlos Criado
|
Mikael Chala, Juan Carlos Criado, Michael Spannowsky
|
The TeraZ Mirage: New Physics Lost in Blind Directions
|
9 pages, 8 figures; clarifications and section V added
| null | null | null |
hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The next generation of high-luminosity electron-positron colliders, such as FCC-ee and CEPC operating at the $Z$ pole (TeraZ), is expected to deliver unprecedented precision in electroweak measurements. These precision observables are typically interpreted within the Standard Model Effective Field Theory (SMEFT), offering a powerful tool to constrain new physics. However, the large number of independent SMEFT operators allows for the possibility of blind directions, parameter combinations to which electroweak precision data are largely insensitive. In this work, we demonstrate that such blind directions are not merely an artefact of agnostic effective field theory scans, but arise generically in realistic ultraviolet completions involving multiple heavy fields. We identify several concrete multi-field extensions of the Standard Model whose low-energy SMEFT projections align with known blind subspaces, and show that these persist even after accounting for renormalisation group evolution and finite one-loop matching effects. Our analysis shows that TeraZ will set a new benchmark in precision for indirect searches, but fully probing the space of possible ultraviolet physics requires moving towards a high energy hadron collider. Later FCC-ee runs at higher centre-of-mass energies, together with the FCC-hh, will provide the necessary complementary probes, enabling a far more complete exploration of the SMEFT parameter space.
|
[
{
"version": "v1",
"created": "Wed, 23 Apr 2025 09:34:07 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 11:05:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chala",
"Mikael",
""
],
[
"Criado",
"Juan Carlos",
""
],
[
"Spannowsky",
"Michael",
""
]
] |
2504.16765
|
Pierre Haas
|
Pierre A. Haas
|
Geometry of T1 transitions in epithelia
|
8 pages, 3 figures; Supplemental Material: 4 pages, 4 figures
|
Phys. Rev. E 112, L042401 (2025)
|
10.1103/kdnb-bxdx
| null |
cond-mat.soft physics.bio-ph
|
http://creativecommons.org/licenses/by/4.0/
|
The flows of tissues of epithelial cells often involve T1 transitions. These neighbour exchanges are irreversible rearrangements crossing an energy barrier. Here, by an exact geometric construction, I determine this energy barrier for general, isolated T1 transitions dominated by line tensions. I~show how deviations from regular cell packing reduce this energy barrier, but find that line tension fluctuations increase it on average. By another exact construction, I prove that the nonlinear tensions in vertex models of tissues also resist T1 transitions. My results thus form the basis for coarse-grained understanding of cell neighbour exchanges for continuum descriptions of epithelia.
|
[
{
"version": "v1",
"created": "Wed, 23 Apr 2025 14:34:34 GMT"
},
{
"version": "v2",
"created": "Mon, 04 Aug 2025 12:39:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Haas",
"Pierre A.",
""
]
] |
2504.16959
|
Xiangyu Zhang
|
Xiangyu Zhang, Erik Floden, Hongru Zhao, Sara Algeri, Galin Jones, Vuk Mandic, Jesse Miller
|
On Validating Angular Power Spectral Models for the Stochastic Gravitational-Wave Background Without Distributional Assumptions
| null | null | null | null |
astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
It is demonstrated that estimators of the angular power spectrum commonly used for the stochastic gravitational-wave background (SGWB) lack a closed-form analytical expression for the likelihood function and, typically, cannot be accurately approximated by a Gaussian likelihood. Nevertheless, a robust statistical analysis can be performed to enable the estimation and testing of angular power spectral models for the SGWB without specifying distributional assumptions. Here, the technical aspects of the method are discussed in detail. Moreover, a new, consistent estimator for the covariance of the angular power spectrum is derived. The proposed approach is applied to data from the third observing run (O3) of Advanced LIGO and Advanced Virgo.
|
[
{
"version": "v1",
"created": "Wed, 23 Apr 2025 05:03:39 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 20:18:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Xiangyu",
""
],
[
"Floden",
"Erik",
""
],
[
"Zhao",
"Hongru",
""
],
[
"Algeri",
"Sara",
""
],
[
"Jones",
"Galin",
""
],
[
"Mandic",
"Vuk",
""
],
[
"Miller",
"Jesse",
""
]
] |
2504.17244
|
Hoang Ly
|
Hoang Ly and Emina Soljanin
|
Service Rate Regions of MDS Codes & Fractional Matchings in Quasi-uniform Hypergraphs
| null | null | null | null |
cs.IT math.CO math.IT
|
http://creativecommons.org/licenses/by/4.0/
|
The service rate region (SRR) has emerged as a critical performance metric for distributed systems that store data redundantly. It measures the system's ability to serve multiple users concurrently. Mathematically, the SRR is a polytope in R^k where each dimension corresponds to the service request rate of one of the k data objects. This paper focuses on systems employing a class of Maximum Distance Separable (MDS) codes. For each code in the class, we characterize the k axes intercept points of its SRR, and the smallest standard simplex that includes the SRR. We use these results to show that the SRR grows with the increasing number of systematic columns in the generator matrices. We establish a graph-theoretic framework associating this SRR problem with fractional matchings in quasi-uniform hypergraphs. Identifying the SRR polytope is equivalent to determining a particular image of the fractional-matching polytope. We introduce a notion of Greedy Matching and show that it is sufficient to focus on these matchings to characterize the SRR rather than the entire matching polytope. With these tools, we determine the SRR of a large subset of the considered class of codes. Our results generalize previous characterizations of systematic and non-systematic MDS-coded systems, offering a unified framework for analyzing service rate regions of codes.
|
[
{
"version": "v1",
"created": "Thu, 24 Apr 2025 04:44:37 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 21:27:27 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ly",
"Hoang",
""
],
[
"Soljanin",
"Emina",
""
]
] |
2504.18382
|
Leonardo Leiva
|
Leonardo Relmucao-Leiva, Carlos Villarroel, Gustavo D\"uring
|
Bridging the Gap Between Avalanche Relaxation and Yielding Rheology
|
Improved organization and readability; content unchanged
| null | null | null |
cond-mat.soft
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The yielding transition in amorphous materials, whether driven passively (simple shear) or actively, remains a fundamental open question in soft matter physics. While avalanche statistics at the critical point have been extensively studied, the emergence of the dynamic regime at yielding and the steady-state flow properties remain poorly understood. In particular, the significant variability observed in flow curves across different systems lacks a clear explanation. We determine, for the first time, the relationship between avalanche duration and size across the yielding transition, revealing how it evolves from quasistatic to dynamic flow regimes. This precise measurement is made using the Controlled Relaxation Time Model (CRTM), a new simulation framework that treats the relaxation time as a tunable parameter. CRTM reproduces known results in both limits and enables a direct analysis of the change of regime between them. Applying the model to different microscopic dynamics, we find that the existing scaling relation connecting critical exponents under flow holds for passive systems. However, active systems exhibit significant deviations, suggesting a missing ingredient in the current understanding of yielding.
|
[
{
"version": "v1",
"created": "Fri, 25 Apr 2025 14:29:59 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:51:46 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Relmucao-Leiva",
"Leonardo",
""
],
[
"Villarroel",
"Carlos",
""
],
[
"Düring",
"Gustavo",
""
]
] |
2504.20068
|
Zhiyu Wu
|
Wei Zhang, Zhiyu Wu, Yi Mu, Ning Rui, Banruo Liu, Nikhil Sarda, Myungjin Lee, Fan Lai
|
JITServe: SLO-aware LLM Serving with Imprecise Request Information
| null | null | null | null |
cs.DC cs.LG cs.SY eess.SY
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
The integration of Large Language Models (LLMs) into applications ranging from interactive chatbots to multi-agent systems has introduced a wide spectrum of service-level objectives (SLOs) for responsiveness. These include latency-sensitive requests emphasizing per-token latency in streaming chat, deadline-sensitive requests requiring rapid full responses to trigger external tools, and compound requests with evolving dependencies across multiple LLM calls. Despite-or perhaps, because of-this workload diversity and unpredictable request information (e.g., response lengths and dependencies), existing request schedulers have focused on aggregate performance, unable to ensure application-level SLO needs.
This paper presents JITServe, the first SLO-aware LLM serving system designed to maximize service goodput (e.g., the number of tokens meeting request SLOs) across diverse workloads. JITServe novelly schedules requests using imprecise request information and gradually relaxes this conservatism by refining request information estimates as generation progresses. It applies a grouped margin goodput maximization algorithm to allocate just enough serving bandwidth to satisfy each request's SLO just-in-time (JIT), maximizing residual capacity for others, while deciding the composition of requests in a batch to maximize efficiency and goodput with provable guarantees. Our evaluation across diverse realistic workloads, including chat, deep research, and agentic pipelines, shows that JITServe improves service goodput by 1.4x-6.3x, alternatively achieving 28.5%-83.2% resource savings, compared to state-of-the-art designs.
|
[
{
"version": "v1",
"created": "Thu, 24 Apr 2025 05:55:21 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 06:24:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Wei",
""
],
[
"Wu",
"Zhiyu",
""
],
[
"Mu",
"Yi",
""
],
[
"Rui",
"Ning",
""
],
[
"Liu",
"Banruo",
""
],
[
"Sarda",
"Nikhil",
""
],
[
"Lee",
"Myungjin",
""
],
[
"Lai",
"Fan",
""
]
] |
2504.20101
|
Yifan Hua
|
Fei Fang, Yifan Hua, Shengze Wang, Ruilin Zhou, Yi Liu, Chen Qian, Xiaoxue Zhang
|
PlanetServe: A Decentralized, Scalable, and Privacy-Preserving Overlay for Democratizing Large Language Model Serving
| null | null | null | null |
cs.DC cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
While significant progress has been made in research and development on open-source and cost-efficient large-language models (LLMs), serving scalability remains a critical challenge, particularly for small organizations and individuals seeking to deploy and test their LLM innovations. Inspired by peer-to-peer networks that leverage decentralized overlay nodes to increase throughput and availability, we propose GenTorrent, an LLM serving overlay that harnesses computing resources from decentralized contributors. We identify four key research problems inherent to enabling such a decentralized infrastructure: 1) overlay network organization; 2) LLM communication privacy; 3) overlay forwarding for resource efficiency; and 4) verification of serving quality. This work presents the first systematic study of these fundamental problems in the context of decentralized LLM serving. Evaluation results from a prototype implemented on a set of decentralized nodes demonstrate that GenTorrent achieves a latency reduction of over 50% compared to the baseline design without overlay forwarding. Furthermore, the security features introduce minimal overhead to serving latency and throughput. We believe this work pioneers a new direction for democratizing and scaling future AI serving capabilities.
|
[
{
"version": "v1",
"created": "Sun, 27 Apr 2025 01:08:25 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Apr 2025 21:24:19 GMT"
},
{
"version": "v3",
"created": "Sat, 30 Aug 2025 16:07:39 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 18:49:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Fang",
"Fei",
""
],
[
"Hua",
"Yifan",
""
],
[
"Wang",
"Shengze",
""
],
[
"Zhou",
"Ruilin",
""
],
[
"Liu",
"Yi",
""
],
[
"Qian",
"Chen",
""
],
[
"Zhang",
"Xiaoxue",
""
]
] |
2504.20543
|
Khun Sang Phukon
|
Khun Sang Phukon, Nathan K. Johnson-McDaniel, Amitesh Singh, Anuradha Gupta
|
Evolution of precessing binary black holes on eccentric orbits using orbit-averaged evolution equations
|
Fixing a typo in Eq. C5c, adding journal information
|
Physical Review D, 112, 044067 (2025)
|
10.1103/2yxp-pcbw, 10.5281/zenodo.16881720.
|
LIGO-P2500139
|
gr-qc astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The most general bound binary black hole (BBH) system has an eccentric orbit and precessing spins. The detection of such a system with significant eccentricity close to the merger would be a clear signature of dynamical formation. In order to study such systems, it is important to be able to evolve their spins and eccentricity from the larger separations at which the binary formed to the smaller separations at which it is detected, or vice versa. Knowledge of the precessional evolution of the binary's orbital angular momentum can also be used to twist up aligned-spin eccentric waveform models to create a spin-precessing eccentric waveform model. In this paper, we present a new publicly available code to evolve eccentric, precessing BBHs using orbit-averaged post-Newtonian (PN) equations from the literature. The spin-precession dynamics is 2PN accurate, i.e., with the leading spin-orbit and spin-spin corrections. The evolution of orbital parameters (orbital frequency, eccentricity, and periastron precession), which follow the quasi-Keplerian parametrization, is 3PN accurate in the point particle terms and includes the leading order spin-orbit and spin-spin effects. All the spin-spin terms include the quadrupole-monopole interaction. The eccentricity enhancement functions in the fluxes use the high-accuracy hyperasymptotic expansions from Loutrel and Yunes [Classical Quantum Gravity {\bf 34} 044003 (2017)]. We discuss various features of the code and study the evolution of the orbital and spin-precession parameters of eccentric, precessing BBHs. In particular, we study the dependence of the spin morphologies on eccentricity, where we find that the transition point from one spin morphology to another can depend nonmonotonically on eccentricity, and the fraction of binaries in a given morphology at a given point in the evolution of a population depends on the instantaneous eccentricity.
|
[
{
"version": "v1",
"created": "Tue, 29 Apr 2025 08:40:07 GMT"
},
{
"version": "v2",
"created": "Mon, 05 May 2025 11:55:56 GMT"
},
{
"version": "v3",
"created": "Mon, 04 Aug 2025 11:57:09 GMT"
},
{
"version": "v4",
"created": "Tue, 02 Sep 2025 13:16:16 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Phukon",
"Khun Sang",
""
],
[
"Johnson-McDaniel",
"Nathan K.",
""
],
[
"Singh",
"Amitesh",
""
],
[
"Gupta",
"Anuradha",
""
]
] |
2504.20566
|
Shunjie Wen
|
Shunjie Wen, Thomas Heinis, Dong-Wan Choi
|
Balanced Online Class-Incremental Learning via Dual Classifiers
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Online class-incremental learning (OCIL) focuses on gradually learning new classes (called plasticity) from a stream of data in a single-pass, while concurrently preserving knowledge of previously learned classes (called stability). The primary challenge in OCIL lies in maintaining a good balance between the knowledge of old and new classes within the continually updated model. Most existing methods rely on explicit knowledge interaction through experience replay, and often employ exclusive training separation to address bias problems. Nevertheless, it still remains a big challenge to achieve a well-balanced learner, as these methods often exhibit either reduced plasticity or limited stability due to difficulties in continually integrating knowledge in the OCIL setting. In this paper, we propose a novel replay-based method, called Balanced Inclusive Separation for Online iNcremental learning (BISON), which can achieve both high plasticity and stability, thus ensuring more balanced performance in OCIL. Our BISON method proposes an inclusive training separation strategy using dual classifiers so that knowledge from both old and new classes can effectively be integrated into the model, while introducing implicit approaches for transferring knowledge across the two classifiers. Extensive experimental evaluations over three widely-used OCIL benchmark datasets demonstrate the superiority of BISON, showing more balanced yet better performance compared to state-of-the-art replay-based OCIL methods.
|
[
{
"version": "v1",
"created": "Tue, 29 Apr 2025 09:13:00 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 05:18:44 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wen",
"Shunjie",
""
],
[
"Heinis",
"Thomas",
""
],
[
"Choi",
"Dong-Wan",
""
]
] |
2504.21336
|
Linshan Wu
|
Linshan Wu, Yuxiang Nie, Sunan He, Jiaxin Zhuang, Luyang Luo, Tao Li, Zhuoyao Xie, Dexuan Chen, Yinghua Zhao, Neeraj Mahboobani, Varut Vardhanabhuti, Ronald Cheong Kin Chan, Yifan Peng, Pranav Rajpurkar, Hao Chen
|
UniBiomed: A Universal Foundation Model for Grounded Biomedical Image Interpretation
|
A universal foundation model for grounded biomedical image interpretation
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The integration of AI-assisted biomedical image analysis into clinical practice demands AI-generated findings that are not only accurate but also interpretable to clinicians. However, existing biomedical AI models generally lack the ability to simultaneously generate diagnostic findings and localize corresponding biomedical objects. This limitation makes it challenging for clinicians to correlate AI-generated findings with visual evidence (e.g., tiny lesions) in images and interpret the results of AI models. To address this challenge, we introduce UniBiomed, the first universal foundation model for grounded biomedical image interpretation, which is capable of generating accurate diagnostic findings and simultaneously segmenting the corresponding biomedical targets. UniBiomed is based on a novel integration of Multi-modal Large Language Model and Segment Anything Model, which can effectively unify diverse biomedical tasks in universal training for advancing grounded interpretation. To develop UniBiomed, we curate a large-scale dataset comprising over 27 million triplets of images, region annotations, and text descriptions across ten biomedical imaging modalities. Extensive validation on 70 internal and 14 external datasets demonstrated the state-of-the-art performance of UniBiomed in diverse biomedical tasks, including image segmentation, disease recognition, region-aware diagnosis, vision question answering, and report generation. In summary, UniBiomed is a powerful and versatile biomedical foundation model, unlocking the untapped grounded interpretation capability for optimizing AI-assisted biomedical image analysis.
|
[
{
"version": "v1",
"created": "Wed, 30 Apr 2025 05:51:48 GMT"
},
{
"version": "v2",
"created": "Thu, 29 May 2025 05:14:48 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 09:08:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wu",
"Linshan",
""
],
[
"Nie",
"Yuxiang",
""
],
[
"He",
"Sunan",
""
],
[
"Zhuang",
"Jiaxin",
""
],
[
"Luo",
"Luyang",
""
],
[
"Li",
"Tao",
""
],
[
"Xie",
"Zhuoyao",
""
],
[
"Chen",
"Dexuan",
""
],
[
"Zhao",
"Yinghua",
""
],
[
"Mahboobani",
"Neeraj",
""
],
[
"Vardhanabhuti",
"Varut",
""
],
[
"Chan",
"Ronald Cheong Kin",
""
],
[
"Peng",
"Yifan",
""
],
[
"Rajpurkar",
"Pranav",
""
],
[
"Chen",
"Hao",
""
]
] |
2504.21349
|
Li Liang
|
Zhenxing Di, Li Liang, Zhiqian Song and Guoliang Tang
|
Gorenstein homological modules over tensor rings
|
Final version, to appear in Kyoto J. Math
| null | null | null |
math.RA math.KT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
For a tensor ring $T_R(M)$, under certain conditions, we characterize the Gorenstein projective modules over $T_R(M)$, and prove that a $T_R(M)$-module $(X,u)$ is Gorenstein projective if and only if $u$ is monomorphic and ${\rm coker}(u)$ is a Gorenstein projective $R$-module. Gorenstein injective (resp., flat) modules over $T_R(M)$ are also explicitly described. Moreover, we give a characterization for the coherence of $T_R(M)$. Some applications to trivial ring extensions and Morita context rings are given.
|
[
{
"version": "v1",
"created": "Wed, 30 Apr 2025 06:17:14 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:40:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Di",
"Zhenxing",
""
],
[
"Liang",
"Li",
""
],
[
"Song",
"Zhiqian",
""
],
[
"Tang",
"Guoliang",
""
]
] |
2504.21638
|
Owen Ekblad
|
Owen Ekblad
|
A note on the quantum Wielandt inequality
|
7 pages; the author was supported in part by US National Science Foundation Grant No. DMS-2153946
| null | null | null |
quant-ph cs.IT math.IT math.OA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this note, we prove that the index of primitivity of any primitive unital Schwarz map is at most $2(D-1)^2$, where $D$ is the dimension of the underlying matrix algebra. This inequality was first proved by Rahaman for Schwarz maps which were both unital and trace preserving. As we show, the assumption of unitality is basically innocuous, but in general not all primitive unital Schwarz maps are trace preserving. Therefore, the precise purpose of this note is to showcase how to apply the method of Rahaman to unital primitive Schwarz maps that don't preserve trace. As a corollary of this theorem, we show that the index of primitivity of any primitive 2-positive map is at most $2(D-1)^2$, so in particular this bound holds for arbitrary primitive completely positive maps. We briefly discuss of how this relates to a conjecture of Perez-Garcia, Verstraete, Wolf and Cirac.
|
[
{
"version": "v1",
"created": "Wed, 30 Apr 2025 13:40:53 GMT"
},
{
"version": "v2",
"created": "Fri, 02 May 2025 20:36:20 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 03:49:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ekblad",
"Owen",
""
]
] |
2505.01730
|
Pranav Ramesh
|
Pranav Ramesh, Gopalakrishnan Srinivasan
|
PASCAL: Precise and Efficient ANN- SNN Conversion using Spike Accumulation and Adaptive Layerwise Activation
| null |
Transactions on Machine Learning Research, 2025
| null | null |
cs.NE cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Spiking Neural Networks (SNNs) have been put forward as an energy-efficient alternative to Artificial Neural Networks (ANNs) since they perform sparse Accumulate operations instead of the power-hungry Multiply-and-Accumulate operations. ANN-SNN conversion is a widely used method to realize deep SNNs with accuracy comparable to that of ANNs.~\citeauthor{bu2023optimal} recently proposed the Quantization-Clip-Floor-Shift (QCFS) activation as an alternative to ReLU to minimize the accuracy loss during ANN-SNN conversion. Nevertheless, SNN inferencing requires a large number of timesteps to match the accuracy of the source ANN for real-world datasets. In this work, we propose PASCAL, which performs ANN-SNN conversion in such a way that the resulting SNN is mathematically equivalent to an ANN with QCFS-activation, thereby yielding similar accuracy as the source ANN with minimal inference timesteps. In addition, we propose a systematic method to configure the quantization step of QCFS activation in a layerwise manner, which effectively determines the optimal number of timesteps per layer for the converted SNN. Our results show that the ResNet-34 SNN obtained using PASCAL achieves an accuracy of $\approx$74\% on ImageNet with a 64$\times$ reduction in the number of inference timesteps compared to existing approaches.
|
[
{
"version": "v1",
"created": "Sat, 03 May 2025 07:55:29 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 13:45:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ramesh",
"Pranav",
""
],
[
"Srinivasan",
"Gopalakrishnan",
""
]
] |
2505.02803
|
Enrico Maria Sessolo
|
Abhishek Chikkaballi, Kamila Kowalska, Rafael R. Lino dos Santos, Enrico Maria Sessolo
|
Stabilizing dark matter with quantum scale symmetry
|
40 pages, 6 figures, 4 tables. Discussion in Sec. 4 extended. References added, typos fixed
|
Phys. Rev. D 112, 115024 (2025)
|
10.1103/2kw8-562l
| null |
hep-ph
|
http://creativecommons.org/licenses/by/4.0/
|
In the context of gauge-Yukawa theories with trans-Planckian asymptotic safety, quantum scale symmetry can prevent the appearance in the Lagrangian of couplings that would otherwise be allowed by the gauge symmetry. Such couplings correspond to irrelevant Gaussian fixed points of the renormalization group flow. Their absence in the theory implies that different sectors of the gauge-Yukawa theory are secluded from one another, in similar fashion to the effects of a global or a discrete symmetry. As an example, we impose the trans-Planckian scale symmetry on a model of Grand Unification based on the gauge group SU(6), showing that it leads to the emergence of several fermionic WIMP dark matter candidates whose coupling strengths are entirely predicted by the UV completion.
|
[
{
"version": "v1",
"created": "Mon, 05 May 2025 17:27:04 GMT"
},
{
"version": "v2",
"created": "Sat, 11 Oct 2025 12:58:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chikkaballi",
"Abhishek",
""
],
[
"Kowalska",
"Kamila",
""
],
[
"Santos",
"Rafael R. Lino dos",
""
],
[
"Sessolo",
"Enrico Maria",
""
]
] |
2505.03212
|
Mingyu Yang
|
Mingyu Yang, Wenxuan Xia, Wentao Li, Raymond Chi-Wing Wong, Wei Wang
|
Elastic Index Selection for Label-Hybrid AKNN Search
|
technical report
| null | null | null |
cs.DB
|
http://creativecommons.org/licenses/by/4.0/
|
Real-world vector embeddings are usually associated with extra labels, such as attributes and keywords. Many applications require the nearest neighbor search that contains specific labels, such as searching for product image embeddings restricted to a particular brand. A straightforward approach is to materialize all possible indices according to the complete query label workload. However, this leads to an exponential increase in both index space and processing time, which significantly limits scalability and efficiency. In this paper, we leverage the inclusion relationships among query label sets to construct partial indexes, enabling index sharing across queries for improved construction efficiency. We introduce \textit{elastic factor} bounds to guarantee search performance and use the greedy algorithm to select indices that meet the bounds, achieving a tradeoff between efficiency and space. Meanwhile, we also designed the algorithm to achieve the best elastic factor under a given space limitation. Experimental results on multiple real datasets demonstrate that our algorithm can achieve near-optimal search performance, achieving up to 10x-500x search efficiency speed up over state-of-the-art approaches. Our algorithm is highly versatile, since it is not constrained by index type and can seamlessly integrate with existing optimized libraries.
|
[
{
"version": "v1",
"created": "Tue, 06 May 2025 05:59:23 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 06:43:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yang",
"Mingyu",
""
],
[
"Xia",
"Wenxuan",
""
],
[
"Li",
"Wentao",
""
],
[
"Wong",
"Raymond Chi-Wing",
""
],
[
"Wang",
"Wei",
""
]
] |
2505.03486
|
Maryam Seifi
|
Maryam Seifi, Zahra Ebadi, Hamzeh Agahi, Hossein Mehri-Dehnavi, Hosein Mohammadzadeh
|
Intrinsic Attractive and Repulsive Interactions: From Classical to Quantum Gases in the Generalized Maxwell-Boltzmann Distribution
|
12 pages,11 figures
|
Physical Review E 111.5 (2025): 054114
|
10.1103/PhysRevE.111.054114
| null |
cond-mat.stat-mech
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The thermodynamic parameter space is flat for an ideal classical gas with non-interacting particles. In contrast, for an ideal quantum Bose (Fermi) gas, the thermodynamic curvature is positive (negative), indicating intrinsic attractive (repulsive) interactions. We generalize the classical Maxwell-Boltzmann distribution by employing a generalized form of the exponential function, proposing the Mittag-Leffler Maxwell-Boltzmann distribution within the framework of superstatistics. We demonstrate that the generalization parameter, $\alpha$, quantifies the statistical interaction. When $\alpha = 1$, the distribution coincides with the standard classical Maxwell-Boltzmann distribution, where no statistical interaction is present. For $0 < \alpha < 1$ ($\alpha > 1$), the statistical interaction is repulsive (attractive), corresponding to a negative (positive) thermodynamic curvature of the system.
|
[
{
"version": "v1",
"created": "Tue, 06 May 2025 12:42:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Seifi",
"Maryam",
""
],
[
"Ebadi",
"Zahra",
""
],
[
"Agahi",
"Hamzeh",
""
],
[
"Mehri-Dehnavi",
"Hossein",
""
],
[
"Mohammadzadeh",
"Hosein",
""
]
] |
2505.03539
|
Kailun Yang
|
Mengfei Duan, Yuheng Zhang, Yihong Cao, Fei Teng, Kai Luo, Jiaming Zhang, Kailun Yang, Zhiyong Li
|
Panoramic Out-of-Distribution Segmentation
|
Code and datasets will be available at https://github.com/MengfeiD/PanOoS
| null | null | null |
cs.CV cs.RO eess.IV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Panoramic imaging enables capturing 360{\deg} images with an ultra-wide Field-of-View (FoV) for dense omnidirectional perception, which is critical to applications, such as autonomous driving and augmented reality, etc. However, current panoramic semantic segmentation methods fail to identify outliers, and pinhole Out-of-distribution Segmentation (OoS) models perform unsatisfactorily in the panoramic domain due to pixel distortions and background clutter. To address these issues, we introduce a new task, Panoramic Out-of-distribution Segmentation (PanOoS), with the aim of achieving comprehensive and safe scene understanding. Furthermore, we propose the first solution, POS, which adapts to the characteristics of panoramic images through text-guided prompt distribution learning. Specifically, POS integrates a disentanglement strategy designed to materialize the cross-domain generalization capability of CLIP. The proposed Prompt-based Restoration Attention (PRA) optimizes semantic decoding by prompt guidance and self-adaptive correction, while Bilevel Prompt Distribution Learning (BPDL) refines the manifold of per-pixel mask embeddings via semantic prototype supervision. Besides, to compensate for the scarcity of PanOoS datasets, we establish two benchmarks: DenseOoS, which features diverse outliers in complex environments, and QuadOoS, captured by a quadruped robot with a panoramic annular lens system. Extensive experiments demonstrate superior performance of POS, with AuPRC improving by 34.25% and FPR95 decreasing by 21.42% on DenseOoS, outperforming state-of-the-art pinhole-OoS methods. Moreover, POS achieves leading closed-set segmentation capabilities and advances the development of panoramic understanding. Code and datasets will be available at https://github.com/MengfeiD/PanOoS.
|
[
{
"version": "v1",
"created": "Tue, 06 May 2025 13:51:26 GMT"
},
{
"version": "v2",
"created": "Fri, 31 Oct 2025 04:10:27 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 14:56:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Duan",
"Mengfei",
""
],
[
"Zhang",
"Yuheng",
""
],
[
"Cao",
"Yihong",
""
],
[
"Teng",
"Fei",
""
],
[
"Luo",
"Kai",
""
],
[
"Zhang",
"Jiaming",
""
],
[
"Yang",
"Kailun",
""
],
[
"Li",
"Zhiyong",
""
]
] |
2505.04638
|
Tianyu Liu
|
Tianyu Liu, Simeng Han, Hanchen Wang, Xiao Luo, Pan Lu, Biqing Zhu, Yuge Wang, Keyi Li, Jiapeng Chen, Rihao Qu, Yufeng Liu, Xinyue Cui, Aviv Yaish, Yuhang Chen, Minsheng Hao, Chuhan Li, Kexing Li, Arman Cohan, Hua Xu, Mark Gerstein, James Zou, Hongyu Zhao
|
Advancing AI Research Assistants with Expert-Involved Learning
|
36 pages, 7 figures
| null | null | null |
cs.AI cs.CL cs.IR
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Large language models (LLMs) and large multimodal models (LMMs) promise to accelerate biomedical discovery, yet their reliability remains unclear. We introduce ARIEL (AI Research Assistant for Expert-in-the-Loop Learning), an open-source evaluation and optimization framework that pairs a curated multimodal biomedical corpus with expert-vetted tasks to probe two capabilities: full-length article summarization and fine-grained figure interpretation. Using uniform protocols and blinded PhD-level evaluation, we find that state-of-the-art models generate fluent but incomplete summaries, whereas LMMs struggle with detailed visual reasoning. We later observe that prompt engineering and lightweight fine-tuning substantially improve textual coverage, and a compute-scaled inference strategy enhances visual question answering. We build an ARIEL agent that integrates textual and visual cues, and we show it can propose testable mechanistic hypotheses. ARIEL delineates current strengths and limitations of foundation models, and provides a reproducible platform for advancing trustworthy AI in biomedicine.
|
[
{
"version": "v1",
"created": "Sat, 03 May 2025 14:21:48 GMT"
},
{
"version": "v2",
"created": "Wed, 08 Oct 2025 23:16:32 GMT"
},
{
"version": "v3",
"created": "Wed, 10 Dec 2025 21:12:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Liu",
"Tianyu",
""
],
[
"Han",
"Simeng",
""
],
[
"Wang",
"Hanchen",
""
],
[
"Luo",
"Xiao",
""
],
[
"Lu",
"Pan",
""
],
[
"Zhu",
"Biqing",
""
],
[
"Wang",
"Yuge",
""
],
[
"Li",
"Keyi",
""
],
[
"Chen",
"Jiapeng",
""
],
[
"Qu",
"Rihao",
""
],
[
"Liu",
"Yufeng",
""
],
[
"Cui",
"Xinyue",
""
],
[
"Yaish",
"Aviv",
""
],
[
"Chen",
"Yuhang",
""
],
[
"Hao",
"Minsheng",
""
],
[
"Li",
"Chuhan",
""
],
[
"Li",
"Kexing",
""
],
[
"Cohan",
"Arman",
""
],
[
"Xu",
"Hua",
""
],
[
"Gerstein",
"Mark",
""
],
[
"Zou",
"James",
""
],
[
"Zhao",
"Hongyu",
""
]
] |
2505.06204
|
Gabriel De Lima Monteiro
|
M. Soledad Aronna, Gabriel de Lima Monteiro, Oscar Sierra Fonseca
|
Average Optimal Control of Uncertain Control-Affine Systems
| null |
Set-Valued Variational Analysis 2025
|
10.1007/s11228-025-00784-0
|
Set-Valued Var. Anal 33, 45
|
math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This work studies optimal control problems of systems with uncertain, probabilistically distributed parameters to optimize average performance. Known as Riemann-Stieltjes, average, or ensemble optimal control, this kind of problem is crucial when parameter uncertainty matters. We derive necessary optimality conditions and characterize feedback controls for control-affine systems. Two scenarios are examined: known initial conditions (finite-dimensional case) and uncertain initial conditions (infinite-dimensional framework). The Pontryagin Maximum Principle is extended using a Hilbert space formulation.
|
[
{
"version": "v1",
"created": "Fri, 09 May 2025 17:30:37 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 12:30:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Aronna",
"M. Soledad",
""
],
[
"Monteiro",
"Gabriel de Lima",
""
],
[
"Fonseca",
"Oscar Sierra",
""
]
] |
2505.06285
|
Yuan Yuhan
|
Yuhan Yuan and Xiaomo Jiang and Haibin Yang and Haixin Zhao and Shengbo Wang and Xueyu Cheng and Jigang Meng and Shuhua Yang
|
FE-MCFormer: An interpretable fault diagnosis framework for rotating machinery under strong noise based on time-frequency fusion transformer
| null | null | null | null |
eess.SP cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Many fault diagnosis methods of rotating machines are based on discriminative features extracted from signals collected from the key components such as bearings. However, under complex operating conditions, periodic impulsive characteristics in the signal related to weak fault information are often obscured by noise interference. Consequently, existing approaches struggle to learn interpretable fault-related features in such scenarios. This paper proposes a novel transformer framework (FE-MCFormer) to extract interpretable time-frequency features, with the aim of improving the fault detection accuracy and intrepretability of rotating machines under strong noise. First, a Fourier adaptive reconstruction embedding layer is introduced as a global information encoder in the model. Subsequently, a time-frequency fusion module is designed, further improve the model robustness and interpretability. The effectiveness of FE-MCFormer in machine fault diagnosis is validated through three case studies.
|
[
{
"version": "v1",
"created": "Wed, 07 May 2025 07:58:48 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 17:26:32 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yuan",
"Yuhan",
""
],
[
"Jiang",
"Xiaomo",
""
],
[
"Yang",
"Haibin",
""
],
[
"Zhao",
"Haixin",
""
],
[
"Wang",
"Shengbo",
""
],
[
"Cheng",
"Xueyu",
""
],
[
"Meng",
"Jigang",
""
],
[
"Yang",
"Shuhua",
""
]
] |
2505.07966
|
Veeti Ahvonen
|
Veeti Ahvonen, Reijo Jaakkola, Antti Kuusisto
|
Formula size game and model checking for modal substitution calculus
| null | null | null | null |
cs.LO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent research has applied modal substitution calculus (MSC) and its variants to characterize various computational frameworks such as graph neural networks (GNNs) and distributed computing systems. For example, it has been shown that the expressive power of recurrent graph neural networks coincides with graded modal substitution calculus GMSC, which is the extension of MSC with counting modalities. GMSC can be further extended with the counting global modality, resulting in the logic GGMSC which corresponds to GNNs with global readout mechanisms. In this paper we introduce a formula-size game that characterizes the expressive power of MSC, GMSC, GGMSC, and related logics. Furthermore, we study the expressiveness and model checking of logics in this family. We prove that MSC and its extensions (GMSC, GGMSC) are as expressive as linear tape-bounded Turing machines, while asynchronous variants are linked to modal mu-calculus and modal computation logic MCL. We establish that for MSC, GMSC and GGMSC, both combined and data complexity of model checking are PSPACE-complete, and for their asynchronous variants, both complexities are PTIME-complete. We also establish that for the propositional fragment SC of MSC, the combined complexity of model checking is PSPACE-complete, while for asynchronous SC it is PTIME-complete, and in both cases, data complexity is constant. As a corollary, we observe that SC satisfiability is PSPACE-complete and NP-complete for its asynchronous variant. Finally, we construct a universal reduction from all recursively enumerable problems to MSC model checking.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 18:07:16 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 14:53:20 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ahvonen",
"Veeti",
""
],
[
"Jaakkola",
"Reijo",
""
],
[
"Kuusisto",
"Antti",
""
]
] |
2505.07979
|
Marie-C\'ecile Piro
|
Daniel Durnford, Yuqi Deng, Carter Garrah, Patrick B. O'Brien, Philippe Gros, Michel Gros, Jos\'e Busto, Steven Kuznicki, Marie-C\'ecile Piro
|
Demonstration of Efficient Radon Removal by Silver-Zeolite in a Dark Matter Detector
|
5 pages
| null | null | null |
hep-ex physics.ins-det
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present the performance of an efficient radon trap using silver-zeolite Ag-ETS-10, measured with a spherical proportional counter filled with an argon/methane mixture. Our study compares the radon reduction capabilities of silver-zeolite and the widely used activated charcoal, both at room temperature. We demonstrate that silver-zeolite significantly outperforms activated charcoal by three orders of magnitude in radon capture. Given that radon is a major background contaminant in rare event searches, our findings highlight silver-zeolite as a highly promising adsorbent, offering compelling operational advantages for both current and future dark matter and neutrino physics experiments. Furthermore, this not only offers great promise for developing future radon reduction systems in underground laboratories, but also paves the way for innovative, multidisciplinary advancements with far-reaching implications in science, engineering and environmental health.
|
[
{
"version": "v1",
"created": "Mon, 12 May 2025 18:23:21 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 17:31:20 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Durnford",
"Daniel",
""
],
[
"Deng",
"Yuqi",
""
],
[
"Garrah",
"Carter",
""
],
[
"O'Brien",
"Patrick B.",
""
],
[
"Gros",
"Philippe",
""
],
[
"Gros",
"Michel",
""
],
[
"Busto",
"José",
""
],
[
"Kuznicki",
"Steven",
""
],
[
"Piro",
"Marie-Cécile",
""
]
] |
2505.08128
|
Changshuai Wei
|
Changshuai Wei, Phuc Nguyen, Benjamin Zelditch, Joyce Chen
|
Beyond Basic A/B testing: Improving Statistical Efficiency for Business Growth
| null | null | null | null |
stat.ME cs.LG math.ST stat.CO stat.TH
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The standard A/B testing approaches are mostly based on t-test in large scale industry applications. These standard approaches however suffers from low statistical power in business settings, due to nature of small sample-size or non-Gaussian distribution or return-on-investment (ROI) consideration. In this paper, we (i) show the statistical efficiency of using estimating equation and U statistics, which can address these issues separately; and (ii) propose a novel doubly robust generalized U that allows flexible definition of treatment effect, and can handles small samples, distribution robustness, ROI and confounding consideration in one framework. We provide theoretical results on asymptotics and efficiency bounds, together with insights on the efficiency gain from theoretical analysis. We further conduct comprehensive simulation studies, apply the methods to multiple real A/B tests at a large SaaS company, and share results and learnings that are broadly useful.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 00:00:06 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 18:04:04 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wei",
"Changshuai",
""
],
[
"Nguyen",
"Phuc",
""
],
[
"Zelditch",
"Benjamin",
""
],
[
"Chen",
"Joyce",
""
]
] |
2505.08218
|
Xin Liang
|
Zhechen Shen and Xin Liang
|
Local Convergence Behavior of Extended LOBPCG for Computing Eigenvalues of Hermitian Matrices
|
25 pages, 9 figures
| null | null | null |
math.NA cs.NA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper provides a comprehensive and detailed analysis of the local convergence behavior of an extended variation of the locally optimal preconditioned conjugate gradient method (LOBPCG) for computing the extreme eigenvalue of a Hermitian matrix. The convergence rates derived in this work are either obtained for the first time or sharper than those previously established, including those in Ovtchinnikov's work ({\em SIAM J. Numer. Anal.}, 46(5):2567--2592, 2008). The study also extends to generalized problems, including Hermitian matrix polynomials that admit an extended form of the Rayleigh quotient. The new approach used to obtain these rates may also serve as a valuable tool for the convergence analysis of other gradient-type optimization methods.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 04:24:09 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Sep 2025 12:43:24 GMT"
},
{
"version": "v3",
"created": "Wed, 10 Dec 2025 15:00:23 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 13:44:36 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Shen",
"Zhechen",
""
],
[
"Liang",
"Xin",
""
]
] |
2505.08530
|
ATLAS Collaboration
|
ATLAS Collaboration
|
The environmental impact, carbon emissions and sustainability of computing in the ATLAS experiment
|
51 pages in total, author list starting page 33, 10 figures, published in EPJC. All figures including auxiliary figures are available at https://atlas.web.cern.ch/Atlas/GROUPS/PHYSICS/PAPERS/SOFT-2024-01
|
Eur. Phys. J. C 85 (2025) 1397
|
10.1140/epjc/s10052-025-14976-3
|
CERN-EP-2025-105
|
hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
ATLAS, a general-purpose experiment at the Large Hadron Collider (LHC), makes use of a large internationally-distributed computing infrastructure, including over $10^6$ TB of managed data on disk and tape and almost one million simultaneously running CPU cores. Upgrades for the High-Luminosity LHC (HL-LHC) will increase the required computing resources by a factor of 3-4 by the beginning of the 2030s, and by an order of magnitude before the conclusion of data taking at the beginning of the 2040s. These resources are spread over around 100 computing sites worldwide. Efforts are underway within the experiment to evaluate and mitigate various aspects of the environmental impact of the sites, with the additional long-term goal of making recommendations to the sites that will significantly reduce the total expected environmental impact in the HL-LHC era. These efforts take several forms: building awareness in the experiment community, adjusting aspects of the computing policy, and modifications of data center configurations, either in ways that take advantage of particular features of ATLAS workloads or in generic ways that reduce the environmental impact of the computing resources. This paper describes the ongoing investigations and approaches that have already provided useful and actionable outcomes.
|
[
{
"version": "v1",
"created": "Tue, 13 May 2025 13:02:24 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:00:45 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"ATLAS Collaboration",
"",
""
]
] |
2505.09202
|
Jan Tristram Acu\~na
|
Jan Tristram Acu\~na, Danny Marfatia, Po-Yan Tseng
|
Angular momentum of vacuum bubbles in a first-order phase transition
|
50 pages, 14 figures, 4 tables. Several improvements
| null | null | null |
hep-ph astro-ph.CO hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
The formation of primordial black holes (PBHs) during a first-order phase transition (FOPT) in a dark sector has been of recent interest. A quantity that characterizes a black hole is its spin. We carry out the first step towards determining the spin of such PBHs, by calculating the spin of spherical false vacuum bubbles induced by cosmological perturbations. The angular momentum is given by the product of density and velocity perturbations. We carefully track the evolution of background quantities and calculate the transfer functions during the FOPT. We find that the dimensionless spin parameter $s = J/(G_{\rm N} M^2)$ of false vacuum bubbles of mass $M$ and angular momentum $J$, take a wide range of values from ${\cal{O}}(10^{-5})$ to ${\cal{O}}(10)$ for FOPTs between 10 keV and 100 GeV and a dark sector that is 0.1 to 0.4 times cooler than the visible sector. We also find a scaling relation between the root-mean-square value of the spin, the FOPT time scale, the bubble wall velocity, and the dark sector-to-visible sector temperature ratio.
|
[
{
"version": "v1",
"created": "Wed, 14 May 2025 07:28:01 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 04:41:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Acuña",
"Jan Tristram",
""
],
[
"Marfatia",
"Danny",
""
],
[
"Tseng",
"Po-Yan",
""
]
] |
2505.10537
|
Noemi Giustini
|
Filippo Olimpieri, Noemi Giustini, Andrea Lacava, Salvatore D'Oro, Tommaso Melodia and Francesca Cuomo
|
LibIQ: Toward Real-Time Spectrum Classification in O-RAN dApps
|
6 pages, 5 figures, 2 tables
|
2025 23rd Mediterranean Communication and Computer Networking Conference (MedComNet), Cagliari, Italy, 2025, pp. 1-6
|
10.1109/MedComNet65822.2025.11100289
| null |
cs.NI cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The O-RAN architecture is transforming cellular networks by adopting RAN softwarization and disaggregation concepts to enable data-driven monitoring and control of the network. Such management is enabled by RICs, which facilitate near-real-time and non-real-time network control through xApps and rApps. However, they face limitations, including latency overhead in data exchange between the RAN and RIC, restricting real-time monitoring, and the inability to access user plain data due to privacy and security constraints, hindering use cases like beamforming and spectrum classification. In this paper, we leverage the dApps concept to enable real-time RF spectrum classification with LibIQ, a novel library for RF signals that facilitates efficient spectrum monitoring and signal classification by providing functionalities to read I/Q samples as time-series, create datasets and visualize time-series data through plots and spectrograms. Thanks to LibIQ, I/Q samples can be efficiently processed to detect external RF signals, which are subsequently classified using a CNN inside the library. To achieve accurate spectrum analysis, we created an extensive dataset of time-series-based I/Q samples, representing distinct signal types captured using a custom dApp running on a 5G deployment over the Colosseum network emulator and an OTA testbed. We evaluate our model by deploying LibIQ in heterogeneous scenarios with varying center frequencies, time windows, and external RF signals. In real-time analysis, the model classifies the processed I/Q samples, achieving an average accuracy of approximately 97.8% in identifying signal types across all scenarios. We pledge to release both LibIQ and the dataset created as a publicly available framework upon acceptance.
|
[
{
"version": "v1",
"created": "Thu, 15 May 2025 17:47:30 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 22:00:27 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 11:18:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Olimpieri",
"Filippo",
""
],
[
"Giustini",
"Noemi",
""
],
[
"Lacava",
"Andrea",
""
],
[
"D'Oro",
"Salvatore",
""
],
[
"Melodia",
"Tommaso",
""
],
[
"Cuomo",
"Francesca",
""
]
] |
2505.11691
|
Karri Koljonen
|
Karri I. I. Koljonen, Manuel Linares
|
SpiderCat: A Catalog of Compact Binary Millisecond Pulsars
|
29 pages, 9 figures, 5 tables. Published in Astrophysical Journal. Updated acknowledgments and bibliography. Publicly accessible web interface at https://astro.phys.ntnu.no/SpiderCAT. Tables 2-5 available in Vizier: https://cdsarc.cds.unistra.fr/viz-bin/cat/J/ApJ/994/8
| null | null | null |
astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present SpiderCat, a multiwavelength catalog of all publicly known compact binary millisecond pulsars (MSPs) in the Galactic field. These systems, colloquially known as "spiders," consist of neutron stars in tight orbits with low-mass companions, which are gradually ablated by the pulsar wind. SpiderCat includes both primary subclasses$-$redbacks and black widows$-$distinguished by companion mass, as well as candidates and peculiar systems such as transitional, huntsman and tidarren MSPs. As of this initial release, SpiderCat contains 111 entries: 30 redbacks, 50 black widows, two huntsmans, 23 redback candidates, five black widow candidates, and one huntsman candidate. In this paper, we compile and summarize key parameters for each system, including spin and orbital properties, and multiwavelength data from radio, optical, X-ray, and $\gamma$-ray observations. An interactive, publicly accessible web interface, at https://astro.phys.ntnu.no/SpiderCAT, enables exploration and visualization of the data. The rapid growth of the number of known spiders, accelerated by the Fermi Large Area Telescope survey and its ability to identify MSPs in $\gamma$-rays, has opened the door to population-level studies. Utilizing SpiderCat, we analyze trends in spin period, orbital period, companion mass, emission properties, and spatial distribution. SpiderCat serves as a dynamic, multiwavelength repository for this unique class of binary pulsars, facilitating new discoveries and constraints on pulsar evolution, particle acceleration, and the neutron star equation of state.
|
[
{
"version": "v1",
"created": "Fri, 16 May 2025 21:00:53 GMT"
},
{
"version": "v2",
"created": "Fri, 05 Sep 2025 17:00:01 GMT"
},
{
"version": "v3",
"created": "Mon, 03 Nov 2025 09:32:38 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 11:45:05 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Koljonen",
"Karri I. I.",
""
],
[
"Linares",
"Manuel",
""
]
] |
2505.11924
|
Yu-Ting Lee
|
Yu-Ting Lee, Fu-Chieh Chang, Hui-Ying Shih, Pei-Yuan Wu
|
Intrinsic Self-Correction in LLMs: Towards Explainable Prompting via Mechanistic Interpretability
| null |
4th Deployable AI Workshop at AAAI 2026
| null | null |
cs.CL cs.AI cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Intrinsic self-correction refers to the phenomenon where a language model refines its own outputs purely through prompting, without external feedback or parameter updates. While this approach improves performance across diverse tasks, its internal mechanism remains poorly understood. We analyze intrinsic self-correction from a representation-level perspective. We formalize and introduce the notion of a prompt-induced shift, which is the change in hidden representations caused by a self-correction prompt. Across 5 open-source LLMs, prompt-induced shifts in text detoxification and text toxification align with latent directions constructed from contrastive pairs. In detoxification, the shifts align with the non-toxic direction; in toxification, they align with the toxic direction. These results suggest that intrinsic self-correction functions as representation steering along interpretable latent directions, beyond what standard metrics such as task scores or model confidence capture. Our analysis offers an interpretability-based account of intrinsic self-correction and contributes to a more systematic understanding of LLM prompting.
|
[
{
"version": "v1",
"created": "Sat, 17 May 2025 09:18:37 GMT"
},
{
"version": "v2",
"created": "Sun, 19 Oct 2025 09:03:49 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Lee",
"Yu-Ting",
""
],
[
"Chang",
"Fu-Chieh",
""
],
[
"Shih",
"Hui-Ying",
""
],
[
"Wu",
"Pei-Yuan",
""
]
] |
2505.12364
|
Francesco Sala
|
Francesco Sala
|
On a Riemann-Roch formula for stacks with finite cyclotomic inertia
| null | null | null | null |
math.AG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
B. Toen defined a Riemann-Roch map from the rational algebraic K-theory of a tame Deligne-Mumford quotient stack to the \'etale K-theory of its inertia. He proved that this map is an isomorphism and that it is covariant with respect to proper maps. Moreover G. Vezzosi and A. Vistoli proved a decomposition theorem for the equivariant K-theory of a noetherian scheme. In this paper we give a geometric definition of the Vezzosi-Vistoli decomposition, interpreting the pieces as corresponding to the components of the cyclotomic inertia. When the map from the cyclotomic inertia to the stack is finite, we can define a Riemann-Roch map in Toen's style. We prove that this map is an isomorphism and it is covariant with respect to proper relatively tame maps; moreover in some favourable circumstances we explicitly compute its inverse map, and show that we can recover Toen's one when the stack is tame Deligne-Mumford.
|
[
{
"version": "v1",
"created": "Sun, 18 May 2025 10:58:20 GMT"
},
{
"version": "v2",
"created": "Wed, 28 May 2025 17:31:53 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 09:58:46 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Sala",
"Francesco",
""
]
] |
2505.12691
|
Haojie Hou
|
Haojie Hou and Yan-Xia Ren and Renming Song
|
Law of iterated logarithm for supercritical non-symmetric branching Markov process
|
49 pages
| null | null | null |
math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
Let $\{(X_t)_{t\geq 0}, \mathbb{P}_{\delta_x}, x\in E\}$ be a supercritical branching Markov process (which is not necessary symmetric) on a locally compact metric measure space $(E,\mu)$ with spatially dependent local branching mechanism. Under some assumptions on the semigroup of the spatial motion, we first prove law of iterated logarithm type results for $\langle f, X_t\rangle$ under the second moment condition on the branching mechanism, where $f$ is a linear combination of eigenfunctions of the mean semigroup $\{T_t, t\geq0\}$ of $X$. Then we prove law of iterated logarithm type results for $\langle f, X_t\rangle$ under the fourth moment condition, where $f$ belongs to a larger class of functions.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 04:30:58 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 14:38:06 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hou",
"Haojie",
""
],
[
"Ren",
"Yan-Xia",
""
],
[
"Song",
"Renming",
""
]
] |
2505.12874
|
Clarissa Buti
|
Lucio Anderlini, Alessandro Bombini, Clarissa Buti, Djunes Janssens, Stefano Lagomarsino, Giovanni Passaleva, Michele Veltri
|
Optimization of 3D diamond detectors with graphitized electrodes based on an innovative numerical simulation
|
29 pages, 18 figures, 2 tables
| null | null | null |
physics.ins-det
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Future experiments at hadron colliders require an evolution of the tracking sensors to ensure sufficient radiation hardness as well as space and time resolution to handle unprecedented particle fluxes. 3D diamond sensors with laser-graphitized electrodes are promising candidates due to their strong binding energy, small atomic number, and high carrier mobility. However, the high resistance of the engraved electrodes delays the propagation of the induced signals towards the readout electronics, thereby degrading the precision of the timing measurements. So far, this effect has been the dominant factor limiting the time resolution of these devices, with other contributions, such as those due to electric field inhomogeneities or electronic noise, typically neglected. Recent advancements in graphitization technology, however, motivate a renewed effort in modeling signal generation in 3D diamond detectors, to achieve more reliable predictions. To this purpose, we apply an extended version of the Ramo-Shockley theorem, describing the effect of signal propagation as a time-dependent weighting potential, obtained by numerically solving the Maxwell's equations in a quasi-static approximation. We developed a custom spectral method solver and validated it against COMSOL MultiPhysics. The response of the modeled sensor to a beam of particles is then simulated using Garfield++ and is compared to the data acquired in a beam test carried on in 2021 by the TimeSPOT Collaboration at the SPS, at CERN. Based on the results obtained with this simulation workflow, we conclude that reducing the resistivity of the graphitic columns remains the priority for significantly improving the time resolution of 3D diamond detectors. Once achieved, optimization of the detector geometry and readout electronics design will become equally important steps to further enhance the timing performance of these devices.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 08:58:39 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Aug 2025 16:13:46 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 10:11:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Anderlini",
"Lucio",
""
],
[
"Bombini",
"Alessandro",
""
],
[
"Buti",
"Clarissa",
""
],
[
"Janssens",
"Djunes",
""
],
[
"Lagomarsino",
"Stefano",
""
],
[
"Passaleva",
"Giovanni",
""
],
[
"Veltri",
"Michele",
""
]
] |
2505.13631
|
Andrei Manolache
|
Andrei Manolache, Luiz F.O. Chamon, Mathias Niepert
|
Learning (Approximately) Equivariant Networks via Constrained Optimization
|
NeurIPS 2025 Oral Camera-Ready
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by-sa/4.0/
|
Equivariant neural networks are designed to respect symmetries through their architecture, boosting generalization and sample efficiency when those symmetries are present in the data distribution. Real-world data, however, often departs from perfect symmetry because of noise, structural variation, measurement bias, or other symmetry-breaking effects. Strictly equivariant models may struggle to fit the data, while unconstrained models lack a principled way to leverage partial symmetries. Even when the data is fully symmetric, enforcing equivariance can hurt training by limiting the model to a restricted region of the parameter space. Guided by homotopy principles, where an optimization problem is solved by gradually transforming a simpler problem into a complex one, we introduce Adaptive Constrained Equivariance (ACE), a constrained optimization approach that starts with a flexible, non-equivariant model and gradually reduces its deviation from equivariance. This gradual tightening smooths training early on and settles the model at a data-driven equilibrium, balancing between equivariance and non-equivariance. Across multiple architectures and tasks, our method consistently improves performance metrics, sample efficiency, and robustness to input perturbations compared with strictly equivariant models and heuristic equivariance relaxations.
|
[
{
"version": "v1",
"created": "Mon, 19 May 2025 18:08:09 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 07:51:20 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Manolache",
"Andrei",
""
],
[
"Chamon",
"Luiz F. O.",
""
],
[
"Niepert",
"Mathias",
""
]
] |
2505.13847
|
Tianle Yang
|
Tianle Yang, Chengzhe Sun, Siwei Lyu, Phil Rose
|
Forensic deepfake audio detection using segmental speech features
|
Accepted for publication in Forensic Science International
| null |
10.1016/j.forsciint.2025.112768
| null |
cs.SD cs.AI cs.CL eess.AS
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This study explores the potential of using acoustic features of segmental speech sounds to detect deepfake audio. These features are highly interpretable because of their close relationship with human articulatory processes and are expected to be more difficult for deepfake models to replicate. The results demonstrate that certain segmental features commonly used in forensic voice comparison (FVC) are effective in identifying deep-fakes, whereas some global features provide little value. These findings underscore the need to approach audio deepfake detection using methods that are distinct from those employed in traditional FVC, and offer a new perspective on leveraging segmental features for this purpose. In addition, the present study proposes a speaker-specific framework for deepfake detection, which differs fundamentally from the speaker-independent systems that dominate current benchmarks. While speaker-independent frameworks aim at broad generalization, the speaker-specific approach offers advantages in forensic contexts where case-by-case interpretability and sensitivity to individual phonetic realization are essential.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 02:42:46 GMT"
},
{
"version": "v2",
"created": "Mon, 02 Jun 2025 02:02:04 GMT"
},
{
"version": "v3",
"created": "Wed, 10 Dec 2025 22:33:52 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yang",
"Tianle",
""
],
[
"Sun",
"Chengzhe",
""
],
[
"Lyu",
"Siwei",
""
],
[
"Rose",
"Phil",
""
]
] |
2505.13886
|
Jingqi Tong
|
Jingqi Tong, Jixin Tang, Hangcheng Li, Yurong Mou, Ming Zhang, Jun Zhao, Yanbo Wen, Fan Song, Jiahao Zhan, Yuyang Lu, Chaoran Tao, Zhiyuan Guo, Jizhou Yu, Tianhao Cheng, Zhiheng Xi, Changhao Jiang, Zhangyue Yin, Yining Zheng, Weifeng Ge, Guanhua Chen, Tao Gui, Xipeng Qiu, Qi Zhang, Xuanjing Huang
|
Game-RL: Synthesizing Multimodal Verifiable Game Data to Boost VLMs' General Reasoning
|
69 pages, 24 figures
| null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Vision-language reinforcement learning (RL) has primarily focused on narrow domains (e.g. geometry or chart reasoning). This leaves broader training scenarios and resources underexplored, limiting the exploration and learning of Vision Language Models (VLMs) through RL. We find video games inherently provide rich visual elements and mechanics that are easy to verify. To fully use the multimodal and verifiable reward in video games, we propose Game-RL, constructing diverse game tasks for RL training to boost VLMs general reasoning ability. To obtain training data, we propose Code2Logic, a novel approach that adapts game code to synthesize game reasoning task data, thus obtaining the GameQA dataset of 30 games and 158 tasks with controllable difficulty gradation. Unexpectedly, RL training solely on GameQA enables multiple VLMs to achieve performance improvements across 7 diverse vision-language benchmarks, demonstrating the value of Game-RL for enhancing VLMs' general reasoning. Furthermore, this suggests that video games may serve as valuable scenarios and resources to boost general reasoning abilities. Our code, dataset and models are available at the GitHub repository.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 03:47:44 GMT"
},
{
"version": "v2",
"created": "Thu, 03 Jul 2025 14:30:25 GMT"
},
{
"version": "v3",
"created": "Thu, 17 Jul 2025 08:46:29 GMT"
},
{
"version": "v4",
"created": "Tue, 16 Sep 2025 15:33:16 GMT"
},
{
"version": "v5",
"created": "Sun, 28 Sep 2025 14:36:59 GMT"
},
{
"version": "v6",
"created": "Sun, 12 Oct 2025 13:28:27 GMT"
},
{
"version": "v7",
"created": "Wed, 03 Dec 2025 11:12:22 GMT"
},
{
"version": "v8",
"created": "Thu, 11 Dec 2025 15:59:47 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tong",
"Jingqi",
""
],
[
"Tang",
"Jixin",
""
],
[
"Li",
"Hangcheng",
""
],
[
"Mou",
"Yurong",
""
],
[
"Zhang",
"Ming",
""
],
[
"Zhao",
"Jun",
""
],
[
"Wen",
"Yanbo",
""
],
[
"Song",
"Fan",
""
],
[
"Zhan",
"Jiahao",
""
],
[
"Lu",
"Yuyang",
""
],
[
"Tao",
"Chaoran",
""
],
[
"Guo",
"Zhiyuan",
""
],
[
"Yu",
"Jizhou",
""
],
[
"Cheng",
"Tianhao",
""
],
[
"Xi",
"Zhiheng",
""
],
[
"Jiang",
"Changhao",
""
],
[
"Yin",
"Zhangyue",
""
],
[
"Zheng",
"Yining",
""
],
[
"Ge",
"Weifeng",
""
],
[
"Chen",
"Guanhua",
""
],
[
"Gui",
"Tao",
""
],
[
"Qiu",
"Xipeng",
""
],
[
"Zhang",
"Qi",
""
],
[
"Huang",
"Xuanjing",
""
]
] |
2505.14134
|
Andrea Mammola
|
Andrea Mammola, Quentin Schaeverbeke, Giuseppe Di Molfetta
|
Noisy simulations of Quantum Walk and Quantum Walk search via Quantum Cellular Automata on a semiconducting spin processor emulator
| null | null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this work we map NISQ-friendly implementations of the non-interacting QCA to a circuit Quantum Electrodynamics (cQED) hardware. We perform both noiseless and noisy simulations of the QCA one particle sector, namely the Quantum Walk, on $N$-cycles and $N \times N$ torus graphs. Moreover, within this framework, we also investigate the search problem and present a circuit for preparing the W state (i.e., the Dicke state with hamming weight one) using only N-1 $\sqrt{\text{iSWAP}}$ gates and no ancilla qubits. The noiseless simulations are conducted with the Qiskit Aer simulator, while the noisy simulations with C12 Quantum Electronics' in-house noisy emulator, \textit{Callisto}. We benchmark the performance of our implementations by analyzing the simulations via relevant metrics and quantities such as the state count distributions, the Hellinger Fidelity, the $\ell^{1}$ distance, the hitting time, and success probability. Our results demonstrate that the QCA framework, in combination with cQED processors, holds promise as an effective platform for early NISQ implementations of Quantum Walk and Quantum Walk Search algorithms.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 09:39:06 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:55:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mammola",
"Andrea",
""
],
[
"Schaeverbeke",
"Quentin",
""
],
[
"Di Molfetta",
"Giuseppe",
""
]
] |
2505.14381
|
Nobuhiro Ueda
|
Yuyang Dong, Nobuhiro Ueda, Kriszti\'an Boros, Daiki Ito, Takuya Sera, Masafumi Oyamada
|
SCAN: Semantic Document Layout Analysis for Textual and Visual Retrieval-Augmented Generation
| null | null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
With the increasing adoption of Large Language Models (LLMs) and Vision-Language Models (VLMs), rich document analysis technologies for applications like Retrieval-Augmented Generation (RAG) and visual RAG are gaining significant attention. Recent research indicates that using VLMs yields better RAG performance, but processing rich documents remains a challenge since a single page contains large amounts of information. In this paper, we present SCAN (SemantiC Document Layout ANalysis), a novel approach that enhances both textual and visual Retrieval-Augmented Generation (RAG) systems that work with visually rich documents. It is a VLM-friendly approach that identifies document components with appropriate semantic granularity, balancing context preservation with processing efficiency. SCAN uses a coarse-grained semantic approach that divides documents into coherent regions covering contiguous components. We trained the SCAN model by fine-tuning object detection models on an annotated dataset. Our experimental results across English and Japanese datasets demonstrate that applying SCAN improves end-to-end textual RAG performance by up to 9.4 points and visual RAG performance by up to 10.4 points, outperforming conventional approaches and even commercial document processing solutions.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 14:03:24 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 08:51:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Dong",
"Yuyang",
""
],
[
"Ueda",
"Nobuhiro",
""
],
[
"Boros",
"Krisztián",
""
],
[
"Ito",
"Daiki",
""
],
[
"Sera",
"Takuya",
""
],
[
"Oyamada",
"Masafumi",
""
]
] |
2505.14527
|
Nitish Shukla
|
Nitish Shukla, Arun Ross
|
diffDemorph: Extending Reference-Free Demorphing to Unseen Faces
| null |
IEEE International Conference on Image Processing (ICIP), 2025
| null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
A face morph is created by combining two face images corresponding to two identities to produce a composite that successfully matches both the constituent identities. Reference-free (RF) demorphing reverses this process using only the morph image, without the need for additional reference images. Previous RF demorphing methods are overly constrained, as they rely on assumptions about the distributions of training and testing morphs such as the morphing technique used (e.g., landmark-based) and face image style (e.g., passport photos). In this paper, we introduce a novel diffusion-based approach, referred to as diffDeMorph, that effectively disentangles component images from a composite morph image with high visual fidelity. Our method is the first to generalize across morph techniques and face styles, beating the current state of the art by $\geq 59.46\%$ under a common training protocol across all datasets tested. We train our method on morphs created using synthetically generated face images and test on real morphs, thereby enhancing the practicality of the technique. Experiments on six datasets and two face matchers establish the utility and efficacy of our method.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 15:48:27 GMT"
},
{
"version": "v2",
"created": "Tue, 27 May 2025 08:05:22 GMT"
},
{
"version": "v3",
"created": "Fri, 06 Jun 2025 13:25:33 GMT"
},
{
"version": "v4",
"created": "Wed, 10 Dec 2025 21:38:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Shukla",
"Nitish",
""
],
[
"Ross",
"Arun",
""
]
] |
2505.14932
|
Isabelle G. Lee
|
Isabelle Lee, Sarah Liaw, Dani Yogatama
|
FOL-Traces: Verified First-Order Logic Reasoning Traces at Scale
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Reasoning in language models is difficult to evaluate: natural-language traces are unverifiable, symbolic datasets too small, and most benchmarks conflate heuristics with inference. We present FOL-Traces, the first large-scale dataset of programmatically verified reasoning traces, enabling rigorous evaluation of structured logical inference. We also propose two challenging and comprehensive diagnostic tasks-masked operation prediction and step completion-that directly probe syntactic awareness and process fidelity. FOL-Traces serves as a scalable testbed for rigorously studying how models perform structured logical inference. Systematic experiments with 5 reasoning LLMs show that the dataset remains challenging: models only reach around 45.7% accuracy on masked operation prediction and around 27% on two-step completion.
|
[
{
"version": "v1",
"created": "Tue, 20 May 2025 21:38:28 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:02:17 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Lee",
"Isabelle",
""
],
[
"Liaw",
"Sarah",
""
],
[
"Yogatama",
"Dani",
""
]
] |
2505.15257
|
Weixiang Zhao
|
Weixiang Zhao, Jiahe Guo, Yang Deng, Tongtong Wu, Wenxuan Zhang, Yulin Hu, Xingyu Sui, Yanyan Zhao, Wanxiang Che, Bing Qin, Tat-Seng Chua, Ting Liu
|
When Less Language is More: Language-Reasoning Disentanglement Makes LLMs Better Multilingual Reasoners
|
NeurIPS 2025 Camera-ready
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Multilingual reasoning remains a significant challenge for large language models (LLMs), with performance disproportionately favoring high-resource languages. Drawing inspiration from cognitive neuroscience, which suggests that human reasoning functions largely independently of language processing, we hypothesize that LLMs similarly encode reasoning and language as separable components that can be disentangled to enhance multilingual reasoning. To evaluate this, we perform a causal intervention by ablating language-specific representations at inference time. Experiments on 10 open-weight LLMs spanning 11 typologically diverse languages show that this language-specific ablation consistently boosts multilingual reasoning performance. Layer-wise analyses further confirm that language and reasoning representations can be effectively disentangled throughout the model, yielding improved multilingual reasoning capabilities, while preserving top-layer language features remains essential for maintaining linguistic fidelity. Compared to post-training methods such as supervised fine-tuning or reinforcement learning, our training-free language-reasoning disentanglement achieves comparable or superior results with minimal computational overhead. These findings shed light on the internal mechanisms underlying multilingual reasoning in LLMs and suggest a lightweight and interpretable strategy for improving cross-lingual generalization.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 08:35:05 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 05:53:45 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhao",
"Weixiang",
""
],
[
"Guo",
"Jiahe",
""
],
[
"Deng",
"Yang",
""
],
[
"Wu",
"Tongtong",
""
],
[
"Zhang",
"Wenxuan",
""
],
[
"Hu",
"Yulin",
""
],
[
"Sui",
"Xingyu",
""
],
[
"Zhao",
"Yanyan",
""
],
[
"Che",
"Wanxiang",
""
],
[
"Qin",
"Bing",
""
],
[
"Chua",
"Tat-Seng",
""
],
[
"Liu",
"Ting",
""
]
] |
2505.15456
|
Weixiang Zhao
|
Weixiang Zhao, Xingyu Sui, Yulin Hu, Jiahe Guo, Haixiao Liu, Biye Li, Yanyan Zhao, Bing Qin, Ting Liu
|
Teaching Language Models to Evolve with Users: Dynamic Profile Modeling for Personalized Alignment
|
NeurIPS 2025 Camera-ready
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Personalized alignment is essential for enabling large language models (LLMs) to engage effectively in user-centric dialogue. While recent prompt-based and offline optimization methods offer preliminary solutions, they fall short in cold-start scenarios and long-term personalization due to their inherently static and shallow designs. In this work, we introduce the Reinforcement Learning for Personalized Alignment (RLPA) framework, in which an LLM interacts with a simulated user model to iteratively infer and refine user profiles through dialogue. The training process is guided by a dual-level reward structure: the Profile Reward encourages accurate construction of user representations, while the Response Reward incentivizes generation of responses consistent with the inferred profile. We instantiate RLPA by fine-tuning Qwen-2.5-3B-Instruct, resulting in Qwen-RLPA, which achieves state-of-the-art performance in personalized dialogue. Empirical evaluations demonstrate that Qwen-RLPA consistently outperforms prompting and offline fine-tuning baselines, and even surpasses advanced commercial models such as Claude-3.5 and GPT-4o. Further analysis highlights Qwen-RLPA's robustness in reconciling conflicting user preferences, sustaining long-term personalization and delivering more efficient inference compared to recent reasoning-focused LLMs. These results emphasize the potential of dynamic profile inference as a more effective paradigm for building personalized dialogue systems.
|
[
{
"version": "v1",
"created": "Wed, 21 May 2025 12:38:36 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 05:49:50 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhao",
"Weixiang",
""
],
[
"Sui",
"Xingyu",
""
],
[
"Hu",
"Yulin",
""
],
[
"Guo",
"Jiahe",
""
],
[
"Liu",
"Haixiao",
""
],
[
"Li",
"Biye",
""
],
[
"Zhao",
"Yanyan",
""
],
[
"Qin",
"Bing",
""
],
[
"Liu",
"Ting",
""
]
] |
2505.17012
|
Haoning Wu
|
Haoning Wu, Xiao Huang, Yaohui Chen, Ya Zhang, Yanfeng Wang, Weidi Xie
|
SpatialScore: Towards Comprehensive Evaluation for Spatial Intelligence
|
Technical Report; Project Page: https://haoningwu3639.github.io/SpatialScore
| null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Existing evaluations of multimodal large language models (MLLMs) on spatial intelligence are typically fragmented and limited in scope. In this work, we aim to conduct a holistic assessment of the spatial understanding capabilities of modern MLLMs and propose complementary data-driven and agent-based solutions. Specifically, we make the following contributions: (i) we introduce SpatialScore, to our knowledge, the most comprehensive and diverse benchmark for multimodal spatial intelligence to date. It covers multiple visual data types, input modalities, and question-answering formats, and contains approximately 5K manually verified samples spanning 30 distinct tasks; (ii) using SpatialScore, we extensively evaluate 40 representative MLLMs, revealing persistent challenges and a substantial gap between current models and human-level spatial intelligence; (iii) to advance model capabilities, we construct SpatialCorpus, a large-scale training resource with 331K multimodal QA samples that supports fine-tuning on spatial reasoning tasks and significantly improves the performance of existing models (e.g., Qwen3-VL); (iv) to complement this data-driven route with a training-free paradigm, we develop SpatialAgent, a multi-agent system equipped with 12 specialized spatial perception tools that supports both Plan-Execute and ReAct reasoning, enabling substantial gains in spatial reasoning without additional model training. Extensive experiments and in-depth analyses demonstrate the effectiveness of our benchmark, corpus, and agent framework. We expect these resources to serve as a solid foundation for advancing MLLMs toward human-level spatial intelligence. All data, code, and models will be released to the research community.
|
[
{
"version": "v1",
"created": "Thu, 22 May 2025 17:59:03 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 13:21:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wu",
"Haoning",
""
],
[
"Huang",
"Xiao",
""
],
[
"Chen",
"Yaohui",
""
],
[
"Zhang",
"Ya",
""
],
[
"Wang",
"Yanfeng",
""
],
[
"Xie",
"Weidi",
""
]
] |
2505.17476
|
Yuchen Zhang
|
Yuchen Zhang, Yaxiong Wang, Yujiao Wu, Lianwei Wu, Li Zhu, Zhedong Zheng
|
The Coherence Trap: When MLLM-Crafted Narratives Exploit Manipulated Visual Contexts
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The detection and grounding of multimedia manipulation has emerged as a critical challenge in combating AI-generated disinformation. While existing methods have made progress in recent years, we identify two fundamental limitations in current approaches: (1) Underestimation of MLLM-driven deception risk: prevailing techniques primarily address rule-based text manipulations, yet fail to account for sophisticated misinformation synthesized by multimodal large language models (MLLMs) that can dynamically generate semantically coherent, contextually plausible yet deceptive narratives conditioned on manipulated images; (2) Unrealistic misalignment artifacts: currently focused scenarios rely on artificially misaligned content that lacks semantic coherence, rendering them easily detectable. To address these gaps holistically, we propose a new adversarial pipeline that leverages MLLMs to generate high-risk disinformation. Our approach begins with constructing the MLLM-Driven Synthetic Multimodal (MDSM) dataset, where images are first altered using state-of-the-art editing techniques and then paired with MLLM-generated deceptive texts that maintain semantic consistency with the visual manipulations. Building upon this foundation, we present the Artifact-aware Manipulation Diagnosis via MLLM (AMD) framework featuring two key innovations: Artifact Pre-perception Encoding strategy and Manipulation-Oriented Reasoning, to tame MLLMs for the MDSM problem. Comprehensive experiments validate our framework's superior generalization capabilities as a unified architecture for detecting MLLM-powered multimodal deceptions. In cross-domain testing on the MDSM dataset, AMD achieves the best average performance, with 88.18 ACC, 60.25 mAP, and 61.02 mIoU scores.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 04:58:27 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 05:36:31 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Yuchen",
""
],
[
"Wang",
"Yaxiong",
""
],
[
"Wu",
"Yujiao",
""
],
[
"Wu",
"Lianwei",
""
],
[
"Zhu",
"Li",
""
],
[
"Zheng",
"Zhedong",
""
]
] |
2505.17508
|
Yifan Zhang
|
Yifan Zhang, Yifeng Liu, Huizhuo Yuan, Yang Yuan, Quanquan Gu, Andrew Chi-Chih Yao
|
On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning
|
Project Page: https://github.com/complex-reasoning/RPG
| null | null | null |
cs.LG cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Policy gradient algorithms have been successfully applied to enhance the reasoning capabilities of large language models (LLMs). KL regularization is ubiquitous, yet the design surface, choice of KL direction (forward vs. reverse), normalization (normalized vs. unnormalized), and estimator ($k_1/k_2/k_3$), is scattered across the literature and often intertwined with off-policy estimation. We ask a focused question: under the off-policy setting, what weighting is required for each KL variant so that the surrogate we optimize yields the exact gradient of the intended KL-regularized objective? We answer this with a compact, unified derivation we call the Regularized Policy Gradient (RPG) view. RPG (i) unifies normalized and unnormalized KL variants and shows that the widely-used $k_3$ penalty is exactly the unnormalized KL; (ii) specifies conditions under which REINFORCE-style losses with stop-gradient are gradient-equivalent to fully differentiable surrogates; (iii) identifies and corrects an off-policy importance-weighting mismatch in GRPO's KL term; and (iv) introduces RPG-Style Clip, a clipped-importance-sampling step within RPG-REINFORCE that enables stable, off-policy policy-gradient training at scale. On mathematical reasoning benchmarks (AIME24, AIME25), RPG-REINFORCE with RPG-Style Clip improves accuracy by up to $+6$ absolute percentage points over DAPO. We extend our experiments to 8K context length, and RPG-REINFORCE with RPG-Style Clip achieves 52% accuracy on AIME25, surpassing the official Qwen3-4B-Instruct model (47%). Notably, RPG is a stable and scalable RL algorithm for LLM reasoning, realized via (a) a KL-correct objective, (b) clipped importance sampling, and (c) an iterative reference-policy update scheme.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 06:01:21 GMT"
},
{
"version": "v2",
"created": "Sun, 28 Sep 2025 23:42:53 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 04:08:28 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Yifan",
""
],
[
"Liu",
"Yifeng",
""
],
[
"Yuan",
"Huizhuo",
""
],
[
"Yuan",
"Yang",
""
],
[
"Gu",
"Quanquan",
""
],
[
"Yao",
"Andrew Chi-Chih",
""
]
] |
2505.17951
|
Haihong Xiao
|
Haihong Xiao and Jianan Zou and Yuxin Zhou and Ying He and Wenxiong Kang
|
SplatCo: Structure-View Collaborative Gaussian Splatting for Detail-Preserving Rendering of Large-Scale Unbounded Scenes
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present SplatCo, a structure-view collaborative Gaussian splatting framework for high-fidelity rendering of complex outdoor scenes. SplatCo builds upon three novel components: 1) a cross-structure collaboration module that combines global tri-plane representations, which capture coarse scene layouts, with local context grid features representing fine details. This fusion is achieved through a hierarchical compensation mechanism, ensuring both global spatial awareness and local detail preservation; 2) a cross-view pruning mechanism that removes overfitted or inaccurate Gaussians based on structural consistency, thereby improving storage efficiency and preventing rendering artifacts; 3) a structure view co-learning module that aggregates structural gradients with view gradients,thereby steering the optimization of Gaussian geometric and appearance attributes more robustly. By combining these key components, SplatCo effectively achieves high-fidelity rendering for large-scale scenes. Code and project page are available at https://splatco-tech.github.io.
|
[
{
"version": "v1",
"created": "Fri, 23 May 2025 14:27:12 GMT"
},
{
"version": "v2",
"created": "Mon, 24 Nov 2025 07:37:20 GMT"
},
{
"version": "v3",
"created": "Tue, 25 Nov 2025 07:30:53 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 06:09:04 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xiao",
"Haihong",
""
],
[
"Zou",
"Jianan",
""
],
[
"Zhou",
"Yuxin",
""
],
[
"He",
"Ying",
""
],
[
"Kang",
"Wenxiong",
""
]
] |
2505.19086
|
Chen Tessler
|
Chen Tessler, Yifeng Jiang, Erwin Coumans, Zhengyi Luo, Gal Chechik, Xue Bin Peng
|
MaskedManipulator: Versatile Whole-Body Manipulation
|
SIGGRAPH Asia 2025 (Project page: https://research.nvidia.com/labs/par/maskedmanipulator/ )
| null | null | null |
cs.RO cs.AI cs.GR
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We tackle the challenges of synthesizing versatile, physically simulated human motions for full-body object manipulation. Unlike prior methods that are focused on detailed motion tracking, trajectory following, or teleoperation, our framework enables users to specify versatile high-level objectives such as target object poses or body poses. To achieve this, we introduce MaskedManipulator, a generative control policy distilled from a tracking controller trained on large-scale human motion capture data. This two-stage learning process allows the system to perform complex interaction behaviors, while providing intuitive user control over both character and object motions. MaskedManipulator produces goal-directed manipulation behaviors that expand the scope of interactive animation systems beyond task-specific solutions.
|
[
{
"version": "v1",
"created": "Sun, 25 May 2025 10:46:14 GMT"
},
{
"version": "v2",
"created": "Sat, 20 Sep 2025 17:43:37 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 17:25:33 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Tessler",
"Chen",
""
],
[
"Jiang",
"Yifeng",
""
],
[
"Coumans",
"Erwin",
""
],
[
"Luo",
"Zhengyi",
""
],
[
"Chechik",
"Gal",
""
],
[
"Peng",
"Xue Bin",
""
]
] |
2505.19486
|
Maonan Wang
|
Maonan Wang, Yirong Chen, Aoyu Pang, Yuxin Cai, Chung Shue Chen, Yuheng Kan, Man-On Pun
|
VLMLight: Safety-Critical Traffic Signal Control via Vision-Language Meta-Control and Dual-Branch Reasoning Architecture
|
25 pages, 15 figures
| null | null | null |
eess.SY cs.LG cs.MA cs.SY
|
http://creativecommons.org/licenses/by/4.0/
|
Traffic signal control (TSC) is a core challenge in urban mobility, where real-time decisions must balance efficiency and safety. Existing methods - ranging from rule-based heuristics to reinforcement learning (RL) - often struggle to generalize to complex, dynamic, and safety-critical scenarios. We introduce VLMLight, a novel TSC framework that integrates vision-language meta-control with dual-branch reasoning. At the core of VLMLight is the first image-based traffic simulator that enables multi-view visual perception at intersections, allowing policies to reason over rich cues such as vehicle type, motion, and spatial density. A large language model (LLM) serves as a safety-prioritized meta-controller, selecting between a fast RL policy for routine traffic and a structured reasoning branch for critical cases. In the latter, multiple LLM agents collaborate to assess traffic phases, prioritize emergency vehicles, and verify rule compliance. Experiments show that VLMLight reduces waiting times for emergency vehicles by up to 65% over RL-only systems, while preserving real-time performance in standard conditions with less than 1% degradation. VLMLight offers a scalable, interpretable, and safety-aware solution for next-generation traffic signal control.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 04:12:57 GMT"
},
{
"version": "v2",
"created": "Fri, 17 Oct 2025 09:23:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wang",
"Maonan",
""
],
[
"Chen",
"Yirong",
""
],
[
"Pang",
"Aoyu",
""
],
[
"Cai",
"Yuxin",
""
],
[
"Chen",
"Chung Shue",
""
],
[
"Kan",
"Yuheng",
""
],
[
"Pun",
"Man-On",
""
]
] |
2505.20257
|
Bo Yang
|
Qianhui Xu, Guangyue Ji, Yuzhu Wang, Ha Quang Trung, Bo Yang
|
Dynamics of Anyon Clusters in Fractional Quantum Hall Fluids
|
16 pages, 13 figures
|
Phys. Rev. B 112, 235112, Editors' Suggestion (2025)
| null | null |
cond-mat.str-el cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In fractional quantum Hall fluids, the quasiparticle excitations are anyons with fractional charges and statistics. Effective interactions among the anyons can be induced by either model or realistic electron-electron (e-e) interactions. Without losing the generality, we investigate such phenomena for the Laughlin 1/3 and Moore-Read (MR) non-Abelian phases. Anyons display rich internal dynamics in both cases that will lead to interesting experimental consequences. In particular, bound states of two Laughlin anyons are preferred under short-range e-e interactions, leading to 2e/3 -- instead of e/3 -- effective charge carriers at low temperatures, which have been seen in several experiments. MR phases host two topologically distinct fusion channels: 1 and psi. The different effective interactions of e/4 anyons in the two sectors suggest the vanishing of the degeneracy of fusion channels when the e-e interaction is no longer its model Hamiltonian, in which case different bound states could also appear. This indicates the possibility of energetically manipulating the two types of anyons by tuning the bare e-e interactions. We point out that the results of recently developed high-resolution STM measurements will be affected by the effective anyon interactions, where anyons are clustered together after the tunneling of electrons. The low-lying parts of the local density of states affected by various anyon clusters are simulated for both Abelian and non-Abelian systems with (screened) Coulomb interactions.
|
[
{
"version": "v1",
"created": "Mon, 26 May 2025 17:36:43 GMT"
},
{
"version": "v2",
"created": "Sat, 01 Nov 2025 02:47:56 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 16:26:27 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Qianhui",
""
],
[
"Ji",
"Guangyue",
""
],
[
"Wang",
"Yuzhu",
""
],
[
"Trung",
"Ha Quang",
""
],
[
"Yang",
"Bo",
""
]
] |
2505.20929
|
Yunhan Du
|
Yunhan Du, Takaaki Aoki, Naoya Fujiwara
|
Potential Landscapes Reveal Spatiotemporal Structure in Urban Mobility: Hodge Decomposition and Principal Component Analysis of Tokyo Before and During COVID-19
| null | null | null | null |
cs.SI physics.soc-ph stat.AP
|
http://creativecommons.org/licenses/by/4.0/
|
Understanding human mobility is vital to solving societal challenges, such as epidemic control and urban transportation optimization. Recent advancements in data collection now enable the exploration of dynamic mobility patterns in human flow. However, the vast volume and complexity of mobility data make it difficult to interpret spatiotemporal patterns directly, necessitating effective information reduction. The core challenge is to balance data simplification with information preservation: methods must retain location-specific information about human flows from origins to destinations while reducing the data to a comprehensible level. This study proposes a two-step dimensionality reduction framework: First, combinatorial Hodge theory is applied to the given origin--destination (OD) matrices with timestamps to construct a set of potential landscapes of human flow, preserving imbalanced trip information between locations. Second, principal component analysis (PCA) expresses the time series of potential landscapes as a linear combination of a few static spatial components, with their coefficients representing temporal variations. The framework systematically decouples the spatial and temporal components of the given data. By implementing this two-step reduction method, we reveal large weight variations during a pandemic, characterized by an overall decline in mobility and stark contrasts between weekdays and holidays. These findings demonstrate the effectiveness of our framework in uncovering complex mobility patterns and its potential to inform urban planning and public health interventions.
|
[
{
"version": "v1",
"created": "Tue, 27 May 2025 09:19:58 GMT"
},
{
"version": "v2",
"created": "Fri, 30 May 2025 04:47:58 GMT"
},
{
"version": "v3",
"created": "Mon, 09 Jun 2025 06:54:28 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 16:15:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Du",
"Yunhan",
""
],
[
"Aoki",
"Takaaki",
""
],
[
"Fujiwara",
"Naoya",
""
]
] |
2505.21356
|
Whenty Ariyanti
|
Whenty Ariyanti, Kuan-Yu Chen, Sabato Marco Siniscalchi, Hsin-Min Wang, Yu Tsao
|
Towards Robust Assessment of Pathological Voices via Combined Low-Level Descriptors and Foundation Model Representations
| null | null | null | null |
cs.SD cs.LG eess.AS
|
http://creativecommons.org/licenses/by/4.0/
|
Perceptual voice quality assessment plays a vital role in diagnosing and monitoring voice disorders. Traditional methods, such as the Consensus Auditory-Perceptual Evaluation of Voice (CAPE-V) and the Grade, Roughness, Breathiness, Asthenia, and Strain (GRBAS) scales, rely on expert raters and are prone to inter-rater variability, emphasizing the need for objective solutions. This study introduces the Voice Quality Assessment Network (VOQANet), a deep learning framework that employs an attention mechanism and Speech Foundation Model (SFM) embeddings to extract high-level features. To further enhance performance, we propose VOQANet+, which integrates self-supervised SFM embeddings with low-level acoustic descriptors-namely jitter, shimmer, and harmonics-to-noise ratio (HNR). Unlike previous approaches that focus solely on vowel-based phonation (PVQD-A), our models are evaluated on both vowel-level and sentence-level speech (PVQD-S) to assess generalizability. Experimental results demonstrate that sentence-based inputs yield higher accuracy, particularly at the patient level. Overall, VOQANet consistently outperforms baseline models in terms of root mean squared error (RMSE) and Pearson correlation coefficient across CAPE-V and GRBAS dimensions, with VOQANet+ achieving even greater performance gains. Additionally, VOQANet+ maintains consistent performance under noisy conditions, suggesting enhanced robustness for real-world and telehealth applications. This work highlights the value of combining SFM embeddings with low-level features for accurate and robust pathological voice assessment.
|
[
{
"version": "v1",
"created": "Tue, 27 May 2025 15:48:17 GMT"
},
{
"version": "v2",
"created": "Wed, 28 May 2025 01:58:07 GMT"
},
{
"version": "v3",
"created": "Fri, 30 May 2025 11:58:33 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 12:26:21 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ariyanti",
"Whenty",
""
],
[
"Chen",
"Kuan-Yu",
""
],
[
"Siniscalchi",
"Sabato Marco",
""
],
[
"Wang",
"Hsin-Min",
""
],
[
"Tsao",
"Yu",
""
]
] |
2505.22234
|
Georgia Kapitsaki
|
Georgia M. Kapitsaki, Maria Papoutsoglou
|
Evolution of repositories and privacy laws: commit activities in the GDPR and CCPA era
| null | null |
10.1016/j.jss.2025.112515
| null |
cs.SE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Free and open source software has gained a lot of momentum in the industry and the research community. The latest advances in privacy legislation, including the EU General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA), have forced the community to pay special attention to users' data privacy. The main aim of this work is to examine software repositories that are acting on privacy laws. We have collected commit data from GitHub repositories in order to understand indications on main data privacy laws (GDPR, CCPA, CPRA, UK DPA) in the last years. Via an automated process, we analyzed 37,213 commits from 12,391 repositories since 2016, whereas 594 commits from the 70 most popular repositories of the dataset were manually analyzed. We observe that most commits were performed on the year the law came into effect and privacy relevant terms appear in the commit messages, whereas reference to specific data privacy user rights is scarce. The study showed that more educational activities on data privacy user rights are needed, as well as tools for privacy recommendations, whereas verifying actual compliance via source code execution is a useful direction for software engineering researchers.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 11:10:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Kapitsaki",
"Georgia M.",
""
],
[
"Papoutsoglou",
"Maria",
""
]
] |
2505.22888
|
Jirui Qi
|
Jirui Qi, Shan Chen, Zidi Xiong, Raquel Fern\'andez, Danielle S. Bitterman, Arianna Bisazza
|
When Models Reason in Your Language: Controlling Thinking Language Comes at the Cost of Accuracy
|
Accepted at EMNLP 2025 Findings
| null |
10.18653/v1/2025.findings-emnlp.1103
| null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Recent Large Reasoning Models (LRMs) with thinking traces have shown strong performance on English reasoning tasks. However, their ability to think in other languages is less studied. This capability is as important as answer accuracy for real world applications because users may find the reasoning trace useful for oversight only when it is expressed in their own language. We comprehensively evaluate two leading families of LRMs on our XReasoning benchmark and find that even the most advanced models often revert to English or produce fragmented reasoning in other languages, revealing a substantial gap in multilingual reasoning. Prompt based interventions that force models to reason in the users language improve readability and oversight but reduce answer accuracy, exposing an important trade off. We further show that targeted post training on just 100 examples mitigates this mismatch, though some accuracy loss remains. Our results highlight the limited multilingual reasoning capabilities of current LRMs and outline directions for future work. Code and data are available at https://github.com/Betswish/mCoT-XReasoning.
|
[
{
"version": "v1",
"created": "Wed, 28 May 2025 21:44:12 GMT"
},
{
"version": "v2",
"created": "Wed, 01 Oct 2025 22:59:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Qi",
"Jirui",
""
],
[
"Chen",
"Shan",
""
],
[
"Xiong",
"Zidi",
""
],
[
"Fernández",
"Raquel",
""
],
[
"Bitterman",
"Danielle S.",
""
],
[
"Bisazza",
"Arianna",
""
]
] |
2506.00693
|
Emory Smith
|
Emory Smith, Robert Wallace, Matthew Robison, Yu Huang, Collin McMillan
|
Human Attention During Localization of Memory Bugs in C Programs
| null | null | null | null |
cs.SE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This paper presents a study of human visual attention during localization of memory bugs in C. Human visual attention refers to the mechanical processes by which we selectively process and prioritize information. Visual attention is important to study because it is central to what information people (who are sighted) use to solve a particular problem. Meanwhile, memory bugs are among the most common types of bugs in C programs that manifest as a variety of program faults. In this paper, we study human visual attention while people attempt to locate memory bugs in code. We recruit 21 programmers to locate between one and eight memory bugs in three C programs for 1.5-2 hours each. In total we collected observations of 31 hours of programmer effort. The bugs in our study cover memory leaks, overflows, and double frees, which are among the most common memory bugs. We analyze the task outcomes in terms of success rate and related factors, patterns of visual attention overall such as what lines and functions are read, and finally we explore differences of visual attention patterns during success versus failure cases.
|
[
{
"version": "v1",
"created": "Sat, 31 May 2025 19:56:16 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 20:52:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Smith",
"Emory",
""
],
[
"Wallace",
"Robert",
""
],
[
"Robison",
"Matthew",
""
],
[
"Huang",
"Yu",
""
],
[
"McMillan",
"Collin",
""
]
] |
2506.01131
|
Yu-Hsueh Chen
|
Yu-Hsueh Chen, Tarun Grover
|
Analogs of deconfined quantum criticality for non-invertible symmetry breaking in 1d
|
16 pages, 4 figures, 1 table; main results are summarized in the Introduction
|
Phys. Rev. B 112, 195129 (2025)
|
10.1103/8tsg-q9j6
| null |
cond-mat.str-el cond-mat.stat-mech hep-th quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
The spontaneous breaking of non-invertible symmetries can lead to exotic phenomena such as coexistence of order and disorder. Here we explore second-order phase transitions in 1d spin chains between two phases that correspond to distinct patterns of non-invertible symmetry breaking. The critical point shares several features with well-understood examples of deconfined quantum critical points, such as enlarged symmetry and identical exponents for the two order parameters participating in the transition. Interestingly, such deconfined transitions involving non-invertible symmetries allow one to construct a whole family of similar critical points by gauging spin-flip symmetries. By employing gauging and bosonization, we characterize the phase diagram of our model in the vicinity of the critical point. We also explore proximate phases and phase transitions in related models, including a deconfined quantum critical point between invertible order parameters that is enforced by a non-invertible symmetry.
|
[
{
"version": "v1",
"created": "Sun, 01 Jun 2025 19:29:44 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Yu-Hsueh",
""
],
[
"Grover",
"Tarun",
""
]
] |
2506.01393
|
Shogo Iwazaki
|
Shogo Iwazaki
|
Improved Regret Bounds for Gaussian Process Upper Confidence Bound in Bayesian Optimization
|
36 pages, NeurIPS 2025
| null | null | null |
cs.LG stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper addresses the Bayesian optimization problem (also referred to as the Bayesian setting of the Gaussian process bandit), where the learner seeks to minimize the regret under a function drawn from a known Gaussian process (GP). Under a Mat\'ern kernel with a certain degree of smoothness, we show that the Gaussian process upper confidence bound (GP-UCB) algorithm achieves $\tilde{O}(\sqrt{T})$ cumulative regret with high probability. Furthermore, our analysis yields $O(\sqrt{T \ln^2 T})$ regret under a squared exponential kernel. These results fill the gap between the existing regret upper bound for GP-UCB and the best-known bound provided by Scarlett (2018). The key idea in our proof is to capture the concentration behavior of the input sequence realized by GP-UCB, enabling a more refined analysis of the GP's information gain.
|
[
{
"version": "v1",
"created": "Mon, 02 Jun 2025 07:38:58 GMT"
},
{
"version": "v2",
"created": "Sun, 13 Jul 2025 05:03:05 GMT"
},
{
"version": "v3",
"created": "Wed, 10 Dec 2025 22:20:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Iwazaki",
"Shogo",
""
]
] |
2506.01524
|
Qi Lin
|
Qi Lin, Weikai Xu, Lisi Chen, Bin Dai
|
V-VAE: A Variational Auto Encoding Framework Towards Fine-Grained Control over Human-Like Chat
| null | null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
With the continued proliferation of Large Language Model (LLM) based chatbots, there is a growing demand for generating responses that are not only linguistically fluent but also consistently aligned with persona-specific traits in conversations. However, existing role-play and persona-based chat approaches rely heavily on static role descriptions, coarse-grained signal space, and low-quality synthetic data, which fail to capture dynamic fine-grained details in human-like chat. Human-like chat requires modeling subtle latent traits, such as emotional tone, situational awareness, and evolving personality, which are difficult to predefine and cannot be easily learned from synthetic or distillation-based data. To address these limitations, we propose a Verbal Variational Auto-Encoding (V-VAE) framework, containing a variational auto-encoding module and fine-grained control space which dynamically adapts dialogue behaviour based on fine-grained, interpretable latent variables across talking style, interaction patterns, and personal attributes. We also construct a high-quality dataset, HumanChatData, and benchmark HumanChatBench to address the scarcity of high-quality data in the human-like domain. Experiments show that LLMs based on V-VAE consistently outperform standard baselines on HumanChatBench and DialogBench, which further demonstrates the effectiveness of V-VAE and HumanChatData.
|
[
{
"version": "v1",
"created": "Mon, 02 Jun 2025 10:38:02 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:22:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Lin",
"Qi",
""
],
[
"Xu",
"Weikai",
""
],
[
"Chen",
"Lisi",
""
],
[
"Dai",
"Bin",
""
]
] |
2506.01625
|
Artun Saday
|
Artun Saday and Ya\c{s}ar Cahit Y{\i}ld{\i}r{\i}m and Cem Tekin
|
Robust Satisficing Gaussian Process Bandits Under Adversarial Attacks
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
We address the problem of Gaussian Process (GP) optimization in the presence of unknown and potentially varying adversarial perturbations. Unlike traditional robust optimization approaches that focus on maximizing performance under worst-case scenarios, we consider a robust satisficing objective, where the goal is to consistently achieve a predefined performance threshold $\tau$, even under adversarial conditions. We propose two novel algorithms based on distinct formulations of robust satisficing, and show that they are instances of a general robust satisficing framework. Further, each algorithm offers different guarantees depending on the nature of the adversary. Specifically, we derive two regret bounds: one that is sublinear over time, assuming certain conditions on the adversary and the satisficing threshold $\tau$, and another that scales with the perturbation magnitude but requires no assumptions on the adversary. Through extensive experiments, we demonstrate that our approach outperforms the established robust optimization methods in achieving the satisficing objective, particularly when the ambiguity set of the robust optimization framework is inaccurately specified.
|
[
{
"version": "v1",
"created": "Mon, 02 Jun 2025 13:04:18 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 23:04:20 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Saday",
"Artun",
""
],
[
"Yıldırım",
"Yaşar Cahit",
""
],
[
"Tekin",
"Cem",
""
]
] |
2506.02130
|
Tyler Cox
|
Tyler A. Cox, Steven G. Murray, Aaron R. Parsons, Joshua S. Dillon, Kartik Mandar, Zachary E. Martinot, Robert Pascua, Piyanat Kittiwisit, and James E. Aguirre
|
fftvis: A Non-Uniform Fast Fourier Transform Based Interferometric Visibility Simulator
|
20 pages, 8 figures, Published to RASTI
| null | null | null |
astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
The detection and characterization of the 21cm signal from the Epoch of Reionization (EoR) demands extraordinary precision in radio interferometric observations and analysis. For modern low-frequency arrays, achieving the dynamic range necessary to detect this signal requires simulation frameworks to validate analysis techniques and characterize systematic effects. However, the computational expense of direct visibility calculations grows rapidly with sky model complexity and array size, posing a potential bottleneck for scalable forward modeling. In this paper, we present fftvis, a high-performance visibility simulator built on the Flatiron Non-Uniform Fast-Fourier Transform (finufft) algorithm. We show that fftvis matches the well-validated matvis simulator to near numerical precision while delivering substantial runtime reductions, up to two orders of magnitude for dense, many-element arrays. We provide a detailed description of the fftvis algorithm and benchmark its computational performance, memory footprint, and numerical accuracy against matvis, including a validation study against analytic solutions for diffuse sky models. We further assess the utility of fftvis in validating 21cm analysis pipelines through a study of the dynamic range in simulated delay and fringe-rate spectra. Our results establish fftvis as a fast, precise, and scalable simulation tool for 21cm cosmology experiments, enabling end-to-end validation of analysis pipelines.
|
[
{
"version": "v1",
"created": "Mon, 02 Jun 2025 18:01:07 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 23:00:24 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Cox",
"Tyler A.",
""
],
[
"Murray",
"Steven G.",
""
],
[
"Parsons",
"Aaron R.",
""
],
[
"Dillon",
"Joshua S.",
""
],
[
"Mandar",
"Kartik",
""
],
[
"Martinot",
"Zachary E.",
""
],
[
"Pascua",
"Robert",
""
],
[
"Kittiwisit",
"Piyanat",
""
],
[
"Aguirre",
"James E.",
""
]
] |
2506.02400
|
Andrew Eberhardt
|
Andrew Eberhardt, Lam Hui
|
Caustic fringes for wave dark matter
| null | null | null | null |
astro-ph.CO astro-ph.GA hep-ph hep-th
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Wave dark matter is composed of particles sufficiently light that their de Broglie wavelength exceeds the average inter-particle separation. A typical wave dark matter halo exhibits granular substructures due to wave interference. In this paper, we explore the wave interference effects around caustics. These are locations of formally divergent density in cold collisionless systems. Examples include splashback in galaxy clusters, and tidal shells in merging galaxies, where the pile-up of dark matter close to apogee gives rise to caustics. We show that wave interference modifies the density profile in the vicinity of the caustics, giving rise to a fringe pattern well-described by the Airy function. This follows from approximating the gravitational potential as linear close to apogee. This prediction is verified in a series of numerical simulations in which the gravitational potential is computed exactly. We provide a formula expressing the fringe separation in terms of the wave dark matter mass and halo parameters, which is useful for interpreting and stacking data. The fringe separation near caustics can be significantly larger than the naive de Broglie scale (the latter set by the system's velocity dispersion). This opens up the possibility of detecting caustic fringes for a wide range of wave dark matter masses.
|
[
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:40:08 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 04:31:56 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Eberhardt",
"Andrew",
""
],
[
"Hui",
"Lam",
""
]
] |
2506.02671
|
Jiazhen Huang
|
Xiao Chen, Jiazhen Huang, Zhiming Liu, Qinting Jiang, Fanding Huang, Jingyan Jiang, Zhi Wang
|
Test-Time Distillation for Continual Model Adaptation
|
11 pages, 6 figures
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Deep neural networks often suffer performance degradation upon deployment due to distribution shifts. Continual Test-Time Adaptation (CTTA) aims to address this issue in an unsupervised manner, yet existing methods, which rely on self-supervision, are prone to an inherent self-referential feedback loop that amplifies initial prediction errors, leading to model drift. We revisit this limitation and propose Test-Time Distillation (TTD), which reframes adaptation as a distillation process guided by a frozen Vision-Language Model (VLM) as an external signal. While promising, we find that direct distillation is fraught with two pitfalls: the Generalist Trap, where the VLM's broad but non-specialized knowledge leads to suboptimal performance on specific tasks and shifts, and the Entropy Bias, where naive model fusion techniques based on entropy fail due to the disparate calibration of heterogeneous models. These pitfalls motivate our insight: the key is to build a robust supervisory signal and leverage it to guide the target model toward stable adaptation. Hence, we present CoDiRe, a Continual Distillation and Rectification framework for TTD. CoDiRe first constructs a robust blended teacher by dynamically fusing the predictions of the VLM and the target model. Critically, it circumvents the Entropy Bias by leveraging Maximum Softmax Probability (MSP) as a more reliable confidence metric for weighting each model's expertise. Then applies an Optimal Transport based rectification to further align predictions with the blended teacher, enabling continuous and stable adaptation. Extensive experiments show that CoDiRe outperforms state-of-the-art baselines, exceeding CoTTA by 10.55% while using only 48% of its time cost on ImageNet-C.
|
[
{
"version": "v1",
"created": "Tue, 03 Jun 2025 09:16:51 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:34:01 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Xiao",
""
],
[
"Huang",
"Jiazhen",
""
],
[
"Liu",
"Zhiming",
""
],
[
"Jiang",
"Qinting",
""
],
[
"Huang",
"Fanding",
""
],
[
"Jiang",
"Jingyan",
""
],
[
"Wang",
"Zhi",
""
]
] |
2506.03712
|
Ruiyang Zhang
|
Ruiyang Zhang, Dongwei Xuan, Jiajun Qin, Lei Zhao, Le Xiao, Xiangming Sun, Lailin Xu, Jianbei Liu
|
Simulation of MAPS and a MAPS-based Inner Tracker for the Super Tau-Charm Facility
|
44 pages, 20 figures, 4 tables, revision submitted to NIM. A
| null | null | null |
physics.ins-det hep-ex
|
http://creativecommons.org/publicdomain/zero/1.0/
|
Monolithic Active Pixel Sensors (MAPS) are a promising detector candidate for the inner tracker of the Super Tau-Charm Facility (STCF). To evaluate the performance of MAPS and the MAPS-based inner tracker, a dedicated simulation workflow has been developed, offering essential insights for detector design and optimization.
The intrinsic characteristics of MAPS, designed using several fabrication processes and pixel geometries, were investigated through a combination of Technology Computer Aided Design (TCAD) and Monte Carlo simulations. Simulations were conducted with both minimum ionizing particles and $^{55}$Fe X-rays to assess critical parameters such as detection efficiency, cluster size, spatial resolution, and charge collection efficiency. Based on these evaluations, a MAPS sensor featuring a strip-like pixel and a high-resistivity epitaxial layer is selected as the baseline sensor design for the STCF inner tracker due to its excellent performance.
Using this optimized MAPS design, a three-layer MAPS-based inner tracker was modeled and simulated. The simulation demonstrated an average detection efficiency exceeding 99%, spatial resolutions of 44.8$\rm{\mu m}$ in the $z$ direction and 8.2$\rm{\mu m}$ in the $r-\phi$ direction, and an intrinsic sensor time resolution of 5.9ns for 1GeV/c $\mu^-$ particles originating from the interaction point. These promising results suggest that the MAPS-based inner tracker fulfills the performance requirements of the STCF experiment.
|
[
{
"version": "v1",
"created": "Wed, 04 Jun 2025 08:44:56 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:55:16 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Ruiyang",
""
],
[
"Xuan",
"Dongwei",
""
],
[
"Qin",
"Jiajun",
""
],
[
"Zhao",
"Lei",
""
],
[
"Xiao",
"Le",
""
],
[
"Sun",
"Xiangming",
""
],
[
"Xu",
"Lailin",
""
],
[
"Liu",
"Jianbei",
""
]
] |
2506.03769
|
Laurent Pedesseau
|
Sreejith Pallikkara Chandrasekharan, Sofia Apergi, Charles Cornet, Laurent Pedesseau
|
Efficient absolute interface energy calculations for heterostructures: Synergy between localized basis sets and surface passivation techniques
|
30 pages, 4 figures
| null | null | null |
cond-mat.mtrl-sci physics.app-ph physics.chem-ph physics.comp-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Heterostructures combining diverse physico-chemical properties are increasingly in demand for a wide range of applications in modern science and technology. However, despite their importance in materials science, accurately determining absolute interface energies remains a major challenge. Here, we present a computationally efficient framework for determining interface energies by incorporating a surface passivation technique, demonstrated using pseudo H passivation with a localized basis set method and an explicit chemical potential. This framework is applied to calculate absolute interface energies and analyze the electronic properties of quasi lattice matched and lattice mismatched III and V on Si interfaces, with results compared to conventional reconstructed surface calculations. By combining localized basis sets with surface passivation techniques, this framework allows for accurate estimation of absolute interface energies in heterogeneous material systems. This approach effectively addresses issues associated with surface reconstructions while significantly reducing computational costs within the framework of density functional theory, and moreover offers considerable potential for calculating interface energies across diverse material systems.
|
[
{
"version": "v1",
"created": "Wed, 04 Jun 2025 09:30:55 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 09:07:02 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chandrasekharan",
"Sreejith Pallikkara",
""
],
[
"Apergi",
"Sofia",
""
],
[
"Cornet",
"Charles",
""
],
[
"Pedesseau",
"Laurent",
""
]
] |
2506.03953
|
Pavlos Lagoudakis Prof
|
Ivan Gnusov, Alexey Yulin, Stepan Baryshev, Sergey Alyatkin, Pavlos G. Lagoudakis
|
Tailoring the resonant spin response of a stirred polariton condensate
| null | null |
10.1103/lxv5-qwxw
| null |
cond-mat.quant-gas cond-mat.mes-hall physics.optics
|
http://creativecommons.org/licenses/by/4.0/
|
We report on the enhancement of the spin coherence time (T2) by almost an order-of-magnitude in exciton-polariton condensates through driven spin precession resonance. Using a rotating optical trap formed by a bichromatic laser excitation, we synchronize the trap stirring frequency with the condensate intrinsic Larmor precession, achieving an order of magnitude increase in spin coherence. By tuning the optical trap profile via excitation lasers intensity, we precisely control the resonance width. Here we present a theoretical model that explains our experimental findings in terms of the mutual synchronization of the condensate circular polarization components. Our findings underpin the potential of polariton condensates for spinoptronic devices and quantum technologies.
|
[
{
"version": "v1",
"created": "Wed, 04 Jun 2025 13:43:58 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Gnusov",
"Ivan",
""
],
[
"Yulin",
"Alexey",
""
],
[
"Baryshev",
"Stepan",
""
],
[
"Alyatkin",
"Sergey",
""
],
[
"Lagoudakis",
"Pavlos G.",
""
]
] |
2506.05191
|
Yake Wei
|
Yake Wei, Yu Miao, Dongzhan Zhou, Di Hu
|
MokA: Multimodal Low-Rank Adaptation for MLLMs
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we reveal that most current efficient multimodal fine-tuning methods are hindered by a key limitation: they are directly borrowed from LLMs, often neglecting the intrinsic differences of multimodal scenarios and even affecting the full utilization of all modalities. Inspired by our empirical observation, we argue that unimodal adaptation and cross-modal adaptation are two essential parts for the effective fine-tuning of MLLMs. From this perspective, we propose Multimodal low-rank Adaptation (MokA), a multimodal-aware efficient fine-tuning strategy that takes multimodal characteristics into consideration. It compresses unimodal information by modality-specific parameters while explicitly enhancing cross-modal interaction, ensuring both unimodal and cross-modal adaptation. Extensive experiments cover three representative multimodal scenarios (audio-visual-text, visual-text, and speech-text), and multiple LLM backbones (LLaMA2/3, Qwen2, Qwen2.5-VL, etc). Consistent improvements indicate the efficacy and versatility of the proposed method. Ablation studies and efficiency evaluation are also conducted to fully asses our method. Overall, we think MokA provides a more targeted solution for efficient adaptation of MLLMs, paving the way for further exploration. The project page is at https://gewu-lab.github.io/MokA.
|
[
{
"version": "v1",
"created": "Thu, 05 Jun 2025 16:04:08 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 12:37:26 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Wei",
"Yake",
""
],
[
"Miao",
"Yu",
""
],
[
"Zhou",
"Dongzhan",
""
],
[
"Hu",
"Di",
""
]
] |
2506.05646
|
Ilgar Baghishov
|
Ilgar Baghishov, Jan Janssen, Graeme Henkelman and Danny Perez
|
Application-specific machine-learned interatomic potentials: exploring the trade-off between DFT convergence, MLIP expressivity, and computational cost
| null |
Digital Discovery, 2026
|
10.1039/d5dd00294j
| null |
physics.comp-ph cond-mat.mtrl-sci
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Machine-learned interatomic potentials (MLIPs) are revolutionizing computational materials science and chemistry by offering an efficient alternative to {\em ab initio} molecular dynamics (MD) simulations. However, fitting high-quality MLIPs remains a challenging, time-consuming, and computationally intensive task where numerous trade-offs have to be considered, e.g., How much and what kind of atomic configurations should be included in the training set? Which level of {\em ab initio} convergence should be used to generate the training set? Which loss function should be used for fitting the MLIP? Which machine learning architecture should be used to train the MLIP? The answers to these questions significantly impact both the computational cost of MLIP training and the accuracy and computational cost of subsequent MLIP MD simulations. In this study, we use a configurationally diverse beryllium dataset and quadratic spectral neighbor analysis potential. We demonstrate that joint optimization of energy versus force weights, training set selection strategies, and convergence settings of the {\em ab initio} reference simulations, as well as model complexity can lead to a significant reduction in the overall computational cost associated with training and evaluating MLIPs. This opens the door to computationally efficient generation of high-quality MLIPs for a range of applications which demand different accuracy versus training and evaluation cost trade-offs.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 00:28:14 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 20:03:46 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Baghishov",
"Ilgar",
""
],
[
"Janssen",
"Jan",
""
],
[
"Henkelman",
"Graeme",
""
],
[
"Perez",
"Danny",
""
]
] |
2506.05848
|
Diego Florez-Ablan
|
Diego Florez-Ablan, Carlos Mejuto-Zaera, Massimo Capone
|
Bath parameterization in multi-band cluster Dynamical Mean-Field Theory
|
17 pages, 10 figures
|
Phys. Rev. B 112, 085153 (2025)
|
10.1103/lnhv-kvy4
| null |
cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Accurate and reliable algorithms to solve complex impurity problems are instrumental to a routine use of quantum embedding methods for material discovery. In this context, we employ an efficient selected configuration interaction impurity solver to investigate the role of bath discretization, specifically, bath size and parameterization, in Hamiltonian-based cluster dynamical mean field theory (CDMFT) for the one- and two-orbital Hubbard models. We consider two- and four-site clusters for the single-orbital model and a two-site cluster for the two-orbital model. Our results demonstrate that, for small bath sizes, the choice of parameterization can significantly influence the solution, highlighting the importance of systematic convergence checks. Comparing different bath parameterizations not only reveals the robustness of a given solution but can also provide insights into the nature of different solutions and potential instabilities of the paramagnetic state. We present an extensive analysis of the zero-temperature Mott transition of the paramagnetic half-filled single-band Hubbard model, benchmarking our findings against previous literature. We find that for the single-band model the dependence on parameterization is weak for the largest bath sizes accessible with ASCI, while a tendency towards a nematic solution can be seen when the bath size is small. Building on this, we extend our study to the multi-band regime, where we present the first systematic analysis at zero temperature for two orbitals and a two-site cluster. This setup allows us to assess the effect of nearest-neighbor dynamical correlations on the multi-orbital Mott transition. In this case, some quantitative dependence on the parameterization is retained for the two-orbital model, for instance in the value of the critical interaction for a Mott transition.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 08:04:40 GMT"
},
{
"version": "v2",
"created": "Fri, 01 Aug 2025 18:32:44 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Florez-Ablan",
"Diego",
""
],
[
"Mejuto-Zaera",
"Carlos",
""
],
[
"Capone",
"Massimo",
""
]
] |
2506.05913
|
Leonie Sch\"urmeyer
|
Leonie Sch\"urmeyer, Ludger Sandig, Jorrit K\"uhne, Leonie Theresa Hezler, Bernd-Wolfgang Igl, Kirsten Schorning
|
Optimal designs for identifying effective doses in drug combination studies
| null | null | null | null |
stat.ME stat.AP
|
http://creativecommons.org/licenses/by/4.0/
|
We consider the optimal design problem for identifying effective dose combinations within drug combination studies where the effect of the combination of two drugs is investigated. Drug combination studies are becoming increasingly important as they investigate potential interaction effects rather than the individual impacts of the drugs. In this situation, identifying effective dose combinations that yield a prespecified effect is of special interest. If nonlinear surface models are used to describe the dose combination-response relationship, these effective dose combinations result in specific contour lines of the fitted response model.
We propose a novel design criterion that targets the precise estimation of these effective dose combinations. In particular, an optimal design minimizes the width of the confidence band of the contour lines of interest. Optimal design theory is developed for this problem, including equivalence theorems and efficiency bounds. The performance of the optimal design is illustrated in different examples modeling dose combination data by various nonlinear surface models. It is demonstrated that the proposed optimal design for identifying effective dose combinations yields a more precise estimation of the effective dose combinations than ray or factorial designs, which are commonly used in practice. This particularly holds true for a case study motivated by data from an oncological dose combination study.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 09:34:33 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 21:17:34 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Schürmeyer",
"Leonie",
""
],
[
"Sandig",
"Ludger",
""
],
[
"Kühne",
"Jorrit",
""
],
[
"Hezler",
"Leonie Theresa",
""
],
[
"Igl",
"Bernd-Wolfgang",
""
],
[
"Schorning",
"Kirsten",
""
]
] |
2506.06158
|
Armand Kassa\"i Koupa\"i
|
Armand Kassa\"i Koupa\"i, Lise Le Boudec, Louis Serrano, Patrick Gallinari
|
ENMA: Tokenwise Autoregression for Generative Neural PDE Operators
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Solving time-dependent parametric partial differential equations (PDEs) remains a fundamental challenge for neural solvers, particularly when generalizing across a wide range of physical parameters and dynamics. When data is uncertain or incomplete-as is often the case-a natural approach is to turn to generative models. We introduce ENMA, a generative neural operator designed to model spatio-temporal dynamics arising from physical phenomena. ENMA predicts future dynamics in a compressed latent space using a generative masked autoregressive transformer trained with flow matching loss, enabling tokenwise generation. Irregularly sampled spatial observations are encoded into uniform latent representations via attention mechanisms and further compressed through a spatio-temporal convolutional encoder. This allows ENMA to perform in-context learning at inference time by conditioning on either past states of the target trajectory or auxiliary context trajectories with similar dynamics. The result is a robust and adaptable framework that generalizes to new PDE regimes and supports one-shot surrogate modeling of time-dependent parametric PDEs.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 15:25:14 GMT"
},
{
"version": "v2",
"created": "Wed, 26 Nov 2025 16:36:22 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 14:55:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Koupaï",
"Armand Kassaï",
""
],
[
"Boudec",
"Lise Le",
""
],
[
"Serrano",
"Louis",
""
],
[
"Gallinari",
"Patrick",
""
]
] |
2506.06277
|
Han Yi
|
Han Yi, Yulu Pan, Feihong He, Xinyu Liu, Benjamin Zhang, Oluwatumininu Oguntola, Gedas Bertasius
|
ExAct: A Video-Language Benchmark for Expert Action Analysis
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
We present ExAct, a new video-language benchmark for expert-level understanding of skilled physical human activities. Our new benchmark contains 3521 expert-curated video question-answer pairs spanning 11 physical activities in 6 domains: Sports, Bike Repair, Cooking, Health, Music, and Dance. ExAct requires the correct answer to be selected from five carefully designed candidate options, thus necessitating a nuanced, fine-grained, expert-level understanding of physical human skills. Evaluating the recent state-of-the-art VLMs on ExAct reveals a substantial performance gap relative to human expert performance. Specifically, the best-performing GPT-4o model achieves only 44.70% accuracy, well below the 82.02% attained by trained human specialists/experts. We believe that ExAct will be beneficial for developing and evaluating VLMs capable of precise understanding of human skills in various physical and procedural domains. Dataset and code are available at https://texaser.github.io/exact_project_page/
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 17:58:51 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 14:46:48 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yi",
"Han",
""
],
[
"Pan",
"Yulu",
""
],
[
"He",
"Feihong",
""
],
[
"Liu",
"Xinyu",
""
],
[
"Zhang",
"Benjamin",
""
],
[
"Oguntola",
"Oluwatumininu",
""
],
[
"Bertasius",
"Gedas",
""
]
] |
2506.06555
|
Morten Hjorth-Jensen
|
Juan Manuel Scarpetta, John Henry Reina, and Morten Hjorth-Jensen
|
Machine learning non-Markovian two-level quantum noise spectroscopy
| null |
Physical Review Research 7, 043285 (2025)
|
10.1103/2lzl-vpjd
| null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We develop machine learning models for the automated characterization of quantum noise spectroscopy for non-Hermitian two-level systems. We use the Random Forest, Support Vector and Feed-Forward Neural Network regression algorithms to perform a highly accurate regression of the two-level system-bath coupling strength. High accuracy Ohmicity classification was implemented to provide a complete characterization of the spectral density function. We define a time-averaged trace-distance metric to feed the machine learning algorithms which, together with numerically exact populations as inputs, produce a highly accurate non-Markovian regression spanning the transition from fast to slow baths and from weak to strong coupling regimes of the interaction. The dynamics database of the non-Hermitian systems has been built up within the independent spin-boson and pure dephasing model.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 22:05:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Scarpetta",
"Juan Manuel",
""
],
[
"Reina",
"John Henry",
""
],
[
"Hjorth-Jensen",
"Morten",
""
]
] |
2506.07811
|
Tieyuan Chen
|
Tieyuan Chen, Huabin Liu, Yi Wang, Chaofan Gan, Mingxi Lyu, Ziran Qin, Shijie Li, Liquan Shen, Junhui Hou, Zheng Wang, Weiyao Lin
|
Looking Beyond Visible Cues: Implicit Video Question Answering via Dual-Clue Reasoning
|
Preprint
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Video Question Answering (VideoQA) aims to answer natural language questions based on the given video, with prior work primarily focusing on identifying the duration of relevant segments, referred to as explicit visual evidence. However, explicit visual evidence is not always directly available, particularly when questions target symbolic meanings or deeper intentions, leading to significant performance degradation. To fill this gap, we introduce a novel task and dataset, $\textbf{I}$mplicit $\textbf{V}$ideo $\textbf{Q}$uestion $\textbf{A}$nswering (I-VQA), which focuses on answering questions in scenarios where explicit visual evidence is inaccessible. Given an implicit question and its corresponding video, I-VQA requires answering based on the contextual visual cues present within the video. To tackle I-VQA, we propose a novel reasoning framework, IRM (Implicit Reasoning Model), incorporating dual-stream modeling of contextual actions and intent clues as implicit reasoning chains. IRM comprises the Action-Intent Module (AIM) and the Visual Enhancement Module (VEM). AIM deduces and preserves question-related dual clues by generating clue candidates and performing relation deduction. VEM enhances contextual visual representation by leveraging key contextual clues. Extensive experiments validate the effectiveness of our IRM in I-VQA tasks, outperforming GPT-4o, OpenAI-o3, and fine-tuned VideoChat2 by $0.76\%$, $1.37\%$, and $4.87\%$, respectively. Additionally, IRM performs SOTA on similar implicit advertisement understanding and future prediction in traffic-VQA. Datasets and codes are available for double-blind review in anonymous repo: https://github.com/tychen-SJTU/Implicit-VideoQA.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 14:38:14 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 07:35:16 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Tieyuan",
""
],
[
"Liu",
"Huabin",
""
],
[
"Wang",
"Yi",
""
],
[
"Gan",
"Chaofan",
""
],
[
"Lyu",
"Mingxi",
""
],
[
"Qin",
"Ziran",
""
],
[
"Li",
"Shijie",
""
],
[
"Shen",
"Liquan",
""
],
[
"Hou",
"Junhui",
""
],
[
"Wang",
"Zheng",
""
],
[
"Lin",
"Weiyao",
""
]
] |
2506.07887
|
Xianjing Dong
|
Xianjing Dong
|
Algebroid Mappings and Their Equidistribution Theory
|
45 pages
| null | null | null |
math.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, the concept of algebroid mappings of complex manifolds is introduced based on that a large number of complex systems of PDEs admit multi-valued solutions that can be defined by a system of independent algebraic equations over the field of meromorphic functions. It is well-known that Nevanlinna theory is an important tool in complex ODE theory. To develop a similar tool applied to the study of algebroid solutions of complex systems of PDEs, one explores the equidistribution theory of algebroid mappings. Via uniformizating an algebroid mapping, we obtain a second main theorem of algebroid mappings from a complete K\"ahler manifold into a complex projective manifold provided that some certain conditions are imposed.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 16:00:17 GMT"
},
{
"version": "v2",
"created": "Tue, 15 Jul 2025 09:20:29 GMT"
},
{
"version": "v3",
"created": "Thu, 18 Sep 2025 10:03:48 GMT"
},
{
"version": "v4",
"created": "Mon, 10 Nov 2025 05:11:56 GMT"
},
{
"version": "v5",
"created": "Thu, 11 Dec 2025 15:18:12 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Dong",
"Xianjing",
""
]
] |
2506.07944
|
The CMS Collaboration
|
CMS Collaboration
|
Determination of the spin and parity of all-charm tetraquarks
|
Replaced with the published version. Added the journal reference and the DOI. All the figures and tables can be found at http://cms-results.web.cern.ch/cms-results/public-results/publications/BPH-24-002 (CMS Public Pages)
|
Nature 648 (2025) 58
|
10.1038/s41586-025-09711-7
|
CMS-BPH-24-002, CERN-EP-2025-118
|
hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
The traditional quark model accounts for the existence of baryons, such as protons and neutrons, which consist of three quarks, as well as mesons, composed of a quark-antiquark pair. Only recently has substantial evidence started to accumulate for exotic states composed of four or five quarks and antiquarks. The exact nature of their internal structure remains uncertain. This paper reports the first measurement of quantum numbers of the recently discovered family of three all-charm tetraquarks, using data collected by the CMS experiment at the Large Hadron Collider from 2016 to 2018. The angular analysis techniques developed for the discovery and characterization of the Higgs boson have been applied to the new exotic states. Here we show that the quantum numbers for parity $P$ and charge conjugation $C$ symmetries are found to be +1. The spin $J$ of these exotic states is consistent with 2$\hbar$, while 0$\hbar$ and 1$\hbar$ are excluded at 95% and 99% confidence level, respectively. The $J^{PC} = 2^{++}$ assignment implies particular configurations of constituent spins and orbital angular momenta, which constrain the possible internal structure of these tetraquarks.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 17:06:06 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 15:34:24 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"CMS Collaboration",
"",
""
]
] |
2506.08001
|
Weiyang Liu
|
Zeju Qiu, Simon Buchholz, Tim Z. Xiao, Maximilian Dax, Bernhard Sch\"olkopf, Weiyang Liu
|
Reparameterized LLM Training via Orthogonal Equivalence Transformation
|
NeurIPS 2025 (40 pages, 26 figures, project page: https://spherelab.ai/poet/, v4: added experiments of finetuning and larger-scale pretraining)
| null | null | null |
cs.LG cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
While large language models (LLMs) are driving the rapid advancement of artificial intelligence, effectively and reliably training these large models remains one of the field's most significant challenges. To address this challenge, we propose POET, a novel reParameterized training algorithm that uses Orthogonal Equivalence Transformation to optimize neurons. Specifically, POET reparameterizes each neuron with two learnable orthogonal matrices and a fixed random weight matrix. Because of its provable preservation of spectral properties of weight matrices, POET can stably optimize the objective function with improved generalization. We further develop efficient approximations that make POET flexible and scalable for training large-scale neural networks. Extensive experiments validate the effectiveness and scalability of POET in training LLMs.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 17:59:34 GMT"
},
{
"version": "v2",
"created": "Sat, 14 Jun 2025 16:13:08 GMT"
},
{
"version": "v3",
"created": "Tue, 17 Jun 2025 16:44:36 GMT"
},
{
"version": "v4",
"created": "Wed, 10 Dec 2025 21:22:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Qiu",
"Zeju",
""
],
[
"Buchholz",
"Simon",
""
],
[
"Xiao",
"Tim Z.",
""
],
[
"Dax",
"Maximilian",
""
],
[
"Schölkopf",
"Bernhard",
""
],
[
"Liu",
"Weiyang",
""
]
] |
2506.08051
|
Shriyank Somvanshi
|
Mahmuda Sultana Mimi, Md Monzurul Islam, Anannya Ghosh Tusti, Shriyank Somvanshi, Subasish Das
|
ST-GraphNet: A Spatio-Temporal Graph Neural Network for Understanding and Predicting Automated Vehicle Crash Severity
| null | null |
10.1145/3764924.3770894
| null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Understanding the spatial and temporal dynamics of automated vehicle (AV) crash severity is critical for advancing urban mobility safety and infrastructure planning. In this work, we introduce ST-GraphNet, a spatio-temporal graph neural network framework designed to model and predict AV crash severity by using both fine-grained and region-aggregated spatial graphs. Using a balanced dataset of 2,352 real-world AV-related crash reports from Texas (2024), including geospatial coordinates, crash timestamps, SAE automation levels, and narrative descriptions, we construct two complementary graph representations: (1) a fine-grained graph with individual crash events as nodes, where edges are defined via spatio-temporal proximity; and (2) a coarse-grained graph where crashes are aggregated into Hexagonal Hierarchical Spatial Indexing (H3)-based spatial cells, connected through hexagonal adjacency. Each node in the graph is enriched with multimodal data, including semantic, spatial, and temporal attributes, including textual embeddings from crash narratives using a pretrained Sentence-BERT model. We evaluate various graph neural network (GNN) architectures, such as Graph Convolutional Networks (GCN), Graph Attention Networks (GAT), and Dynamic Spatio-Temporal GCN (DSTGCN), to classify crash severity and predict high-risk regions. Our proposed ST-GraphNet, which utilizes a DSTGCN backbone on the coarse-grained H3 graph, achieves a test accuracy of 97.74\%, substantially outperforming the best fine-grained model (64.7\% test accuracy). These findings highlight the effectiveness of spatial aggregation, dynamic message passing, and multi-modal feature integration in capturing the complex spatio-temporal patterns underlying AV crash severity.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 01:42:19 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Mimi",
"Mahmuda Sultana",
""
],
[
"Islam",
"Md Monzurul",
""
],
[
"Tusti",
"Anannya Ghosh",
""
],
[
"Somvanshi",
"Shriyank",
""
],
[
"Das",
"Subasish",
""
]
] |
2506.09029
|
Josias Old
|
Josias Old, Stephan Tasler, Michael J. Hartmann, Markus M\"uller
|
Fault-Tolerant Stabilizer Measurements in Surface Codes with Three-Qubit Gates
|
7 pages, 6 figures
| null |
10.1103/sblg-fbq4
| null |
quant-ph
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Quantum error correction (QEC) is considered a deciding component in enabling practical quantum computing. Stabilizer codes, and in particular topological surface codes, are promising candidates for implementing QEC by redundantly encoding quantum information. While it is widely believed that a strictly fault-tolerant protocol can only be implemented using single- and two-qubit gates, several quantum computing platforms, based on trapped ions, neutral atoms and also superconducting qubits support native multi-qubit operations, e.g. using multi-ion entangling gates, Rydberg blockade or parallelized tunable couplers, respectively. In this work, we show that stabilizer measurement circuits for unrotated surface codes can be fault-tolerant using single auxiliary qubits and three-qubit gates. These gates enable lower-depth circuits leading to fewer fault locations and potentially shorter QEC cycle times. Concretely, we find that in an optimistic parameter regime where fidelities of three-qubit gates are the same as those of two-qubit gates, the logical error rate can be up to one order of magnitude lower and the threshold can be significantly higher, increasing from $\approx 0.76 \%$ to $\approx 1.05 \%$. Our results, which are applicable to a wide range of platforms, thereby motivate further investigation into multi-qubit gates as components for fault-tolerant QEC, as they can lead to substantial advantages in terms of time and physical qubit resources required to reach a target logical error rate.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 17:54:23 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Old",
"Josias",
""
],
[
"Tasler",
"Stephan",
""
],
[
"Hartmann",
"Michael J.",
""
],
[
"Müller",
"Markus",
""
]
] |
2506.09664
|
Pascal Michaillat
|
Pascal Michaillat
|
Recession Detection Using Classifiers on the Anticipation-Precision Frontier
| null | null | null | null |
econ.GN q-fin.EC
|
http://creativecommons.org/licenses/by/4.0/
|
This paper develops an algorithm for detecting US recessions in real time. The algorithm constructs hundreds of millions of recession classifiers by combining unemployment and vacancy data. Classifiers are then selected to avoid both false negatives (missed recessions) and false positives (nonexistent recessions). The selected classifiers are perfect in a statistical sense: they identify all 15 historical recessions in the 1929--2021 training period without any false positives. By further selecting classifiers that lie on the high-precision segment of the anticipation-precision frontier, the algorithm delivers early detection without sacrificing accuracy. On average between 1929 and 2021, the selected classifier ensemble signals recessions 2.1 months after their true onset, with a standard deviation of detection errors of 1.8 months. The classifier ensemble is much faster than the NBER Business Cycle Dating Committee: between 1979 and 2021, the committee takes on average 6.3 months to determine recession starts, while the classifier ensemble only takes 1.2 months. Applied to September 2025 data, the classifier ensemble gives a 64% probability that the US economy has entered a recession. A placebo test and backtests confirm the algorithm's reliability.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 12:35:22 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 14:47:20 GMT"
},
{
"version": "v3",
"created": "Mon, 07 Jul 2025 15:57:11 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 14:07:25 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Michaillat",
"Pascal",
""
]
] |
2506.09881
|
Siyu Chen
|
Siyu Chen, Ting Han, Chengzheng Fu, Changshe Zhang, Chaolei Wang, Jinhe Su, Guorong Cai, Meiliu Wu
|
Leveraging Depth and Language for Open-Vocabulary Domain-Generalized Semantic Segmentation
|
Accepted by NeurIPS 2025
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Open-Vocabulary semantic segmentation (OVSS) and domain generalization in semantic segmentation (DGSS) highlight a subtle complementarity that motivates Open-Vocabulary Domain-Generalized Semantic Segmentation (OV-DGSS). OV-DGSS aims to generate pixel-level masks for unseen categories while maintaining robustness across unseen domains, a critical capability for real-world scenarios such as autonomous driving in adverse conditions. We introduce Vireo, a novel single-stage framework for OV-DGSS that unifies the strengths of OVSS and DGSS for the first time. Vireo builds upon the frozen Visual Foundation Models (VFMs) and incorporates scene geometry via Depth VFMs to extract domain-invariant structural features. To bridge the gap between visual and textual modalities under domain shift, we propose three key components: (1) GeoText Prompts, which align geometric features with language cues and progressively refine VFM encoder representations; (2) Coarse Mask Prior Embedding (CMPE) for enhancing gradient flow for faster convergence and stronger textual influence; and (3) the Domain-Open-Vocabulary Vector Embedding Head (DOV-VEH), which fuses refined structural and semantic features for robust prediction. Comprehensive evaluation on these components demonstrates the effectiveness of our designs. Our proposed Vireo achieves the state-of-the-art performance and surpasses existing methods by a large margin in both domain generalization and open-vocabulary recognition, offering a unified and scalable solution for robust visual understanding in diverse and dynamic environments. Code is available at https://github.com/anonymouse-9c53tp182bvz/Vireo.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 15:54:47 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 13:58:29 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 14:33:57 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Siyu",
""
],
[
"Han",
"Ting",
""
],
[
"Fu",
"Chengzheng",
""
],
[
"Zhang",
"Changshe",
""
],
[
"Wang",
"Chaolei",
""
],
[
"Su",
"Jinhe",
""
],
[
"Cai",
"Guorong",
""
],
[
"Wu",
"Meiliu",
""
]
] |
2506.10177
|
Defang Chen Dr.
|
Defang Chen, Zhenyu Zhou, Can Wang, Siwei Lyu
|
Geometric Regularity in Deterministic Sampling Dynamics of Diffusion-based Generative Models
|
57 pages. Accepted by Journal of Statistical Mechanics: Theory and Experiment (2025). The short version was published in ICML 2024. arXiv admin note: text overlap with arXiv:2405.11326
|
J. Stat. Mech. (2025) 124002
|
10.1088/1742-5468/ae17ac
| null |
cs.LG cond-mat.stat-mech cs.CV stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Diffusion-based generative models employ stochastic differential equations (SDEs) and their equivalent probability flow ordinary differential equations (ODEs) to establish a smooth transformation between complex high-dimensional data distributions and tractable prior distributions. In this paper, we reveal a striking geometric regularity in the deterministic sampling dynamics of diffusion generative models: each simulated sampling trajectory along the gradient field lies within an extremely low-dimensional subspace, and all trajectories exhibit an almost identical boomerang shape, regardless of the model architecture, applied conditions, or generated content. We characterize several intriguing properties of these trajectories, particularly under closed-form solutions based on kernel-estimated data modeling. We also demonstrate a practical application of the discovered trajectory regularity by proposing a dynamic programming-based scheme to better align the sampling time schedule with the underlying trajectory structure. This simple strategy requires minimal modification to existing deterministic numerical solvers, incurs negligible computational overhead, and achieves superior image generation performance, especially in regions with only 5 - 10 function evaluations.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 21:09:09 GMT"
},
{
"version": "v2",
"created": "Thu, 27 Nov 2025 14:56:01 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 04:38:09 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Defang",
""
],
[
"Zhou",
"Zhenyu",
""
],
[
"Wang",
"Can",
""
],
[
"Lyu",
"Siwei",
""
]
] |
2506.10187
|
Denis Novokreschenov
|
D. Novokreschenov, V. Neplokh, M. Misko, N. Starkova, T. Cookson, A. Kudlis, A. Nalitov, I. A. Shelykh, A. V. Kavokin, and P. Lagoudakis
|
Kardar-Parisi-Zhang universality in optically induced lattices of exciton-polariton condensates
| null | null | null | null |
cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate space-time coherence in one-dimensional lattices of exciton-polariton condensates formed by fully reconfigurable non-resonant optical pumping. Starting from an open-dissipative Gross-Pitaevskii equation with deterministic reservoir kinetics and stochastic condensate noise, we derive a discrete complex-field model that incorporates coherent tunnelling, reservoir-mediated dissipative coupling and gain-saturation non-linearity. Adiabatic elimination of fast density fluctuations reveals a wedge-shaped region in the complex hopping plane where the coarse-grained phase dynamics reduces to the Kardar-Parisi-Zhang (KPZ) equation. By computing high-resolution phase diagrams of the temporal and spatial scaling exponents we pinpoint the boundaries separating the KPZ domain from the Edwards-Wilkinson (EW) regime. Large-scale graphics processing unit (GPU) simulations of chains containing up to $N=2000$ condensates confirm these predictions: inside the wedge the exponents converge to $\beta_{N}=\textbf{0.329}(3)\!\approx\!1/3$ and $\chi_{N}=\textbf{0.504}(4)\!\approx\!1/2$, whereas outside it the dynamics moves away from KPZ and ultimately flows toward the EW fixed point, although finite system size and finite observation time may yield intermediate effective exponents. These results pave the way to the implementation of ultrafast KPZ-simulators based on one-dimensional arrays of exciton-polariton condensates.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 21:25:30 GMT"
},
{
"version": "v2",
"created": "Fri, 13 Jun 2025 09:23:46 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 18:54:10 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Novokreschenov",
"D.",
""
],
[
"Neplokh",
"V.",
""
],
[
"Misko",
"M.",
""
],
[
"Starkova",
"N.",
""
],
[
"Cookson",
"T.",
""
],
[
"Kudlis",
"A.",
""
],
[
"Nalitov",
"A.",
""
],
[
"Shelykh",
"I. A.",
""
],
[
"Kavokin",
"A. V.",
""
],
[
"Lagoudakis",
"P.",
""
]
] |
2506.10473
|
Tristan Bullion-Gauthier
|
Tristan Bullion-Gauthier (ICJ, EDPA)
|
Higher-order affine Sobolev inequalities
|
New result added. Some typos corrected
| null | null | null |
math.FA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Zhang refined the classical Sobolev inequality $\|f\|_{L^{Np/(N-p)}} \lesssim \| \nabla f \|_{L^p}$, where $1\leq p \lt N$, by replacing $\|\nabla f\|_{L^p}$ with a smaller quantity invariant by unimodular affine transformations. The analogue result in homogeneous fractional Sobolev spaces $\mathring{W}^{s,p}$, with $0 \lt s \lt 1$ and $sp \lt N$, was obtained by Haddad and Ludwig. We generalize their results to the case where $s \gt 1$. Our approach, based on the existence of optimal unimodular transformations, allows us to obtain various affine inequalities, such as affine Sobolev inequalities, reverse affine inequalities, and affine Gagliardo-Nirenberg type inequalities. In a different but related direction, we also answer a question concerning reverse affine inequalities, raised by Haddad, Jim\'enez, and Montenegro.
|
[
{
"version": "v1",
"created": "Thu, 12 Jun 2025 08:25:49 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 21:18:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Bullion-Gauthier",
"Tristan",
"",
"ICJ, EDPA"
]
] |
2506.11213
|
Matt Booth
|
Matt Booth, Isambard Goodbody, Sebastian Opper
|
Reflexive dg categories in algebra and topology
|
43 pages, 2 figures. Comments welcome! v2: some results generalised, exposition improved
| null | null | null |
math.RT math.AG math.AT math.CT math.SG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reflexive dg categories were introduced by Kuznetsov and Shinder to abstract the duality between bounded and perfect derived categories. In particular this duality relates their Hochschild cohomologies, autoequivalence groups, and semiorthogonal decompositions. We establish reflexivity in a variety of settings including affine schemes, simple-minded collections, chain and cochain dg algebras of topological spaces, Ginzburg dg algebras, and Fukaya categories of cotangent bundles and surfaces as well as the closely related class of graded gentle algebras. Our proofs are based on the interplay of reflexivity with gluings, derived completions, and Koszul duality. In particular we show that for certain (co)connective dg algebras, reflexivity is equivalent to derived completeness.
|
[
{
"version": "v1",
"created": "Thu, 12 Jun 2025 18:20:28 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:25:28 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Booth",
"Matt",
""
],
[
"Goodbody",
"Isambard",
""
],
[
"Opper",
"Sebastian",
""
]
] |
2506.11375
|
Yitong Zhou
|
Yitong Zhou, Mingyue Cheng, Qingyang Mao, Yucong Luo, Qi Liu, Yupeng Li, Xiaohan Zhang, Deguang Liu, Xin Li, Enhong Chen
|
Benchmarking Multimodal LLMs on Recognition and Understanding over Chemical Tables
| null | null | null | null |
cs.AI cs.CL
|
http://creativecommons.org/licenses/by-sa/4.0/
|
With the widespread application of multimodal large language models in scientific intelligence, there is an urgent need for more challenging evaluation benchmarks to assess their ability to understand complex scientific data. Scientific tables, as core carriers of knowledge representation, combine text, symbols, and graphics, forming a typical multimodal reasoning scenario. However, existing benchmarks are mostly focused on general domains, failing to reflect the unique structural complexity and domain-specific semantics inherent in scientific research. Chemical tables are particularly representative: they intertwine structured variables such as reagents, conditions, and yields with visual symbols like molecular structures and chemical formulas, posing significant challenges to models in cross-modal alignment and semantic parsing. To address this, we propose ChemTable-a large scale benchmark of chemical tables constructed from real-world literature, containing expert-annotated cell layouts, logical structures, and domain-specific labels. It supports two core tasks: (1) table recognition (structure and content extraction); and (2) table understanding (descriptive and reasoning-based question answering). Evaluation on ChemTable shows that while mainstream multimodal models perform reasonably well in layout parsing, they still face significant limitations when handling critical elements such as molecular structures and symbolic conventions. Closed-source models lead overall but still fall short of human-level performance. This work provides a realistic testing platform for evaluating scientific multimodal understanding, revealing the current bottlenecks in domain-specific reasoning and advancing the development of intelligent systems for scientific research.
|
[
{
"version": "v1",
"created": "Fri, 13 Jun 2025 00:45:41 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 02:34:29 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhou",
"Yitong",
""
],
[
"Cheng",
"Mingyue",
""
],
[
"Mao",
"Qingyang",
""
],
[
"Luo",
"Yucong",
""
],
[
"Liu",
"Qi",
""
],
[
"Li",
"Yupeng",
""
],
[
"Zhang",
"Xiaohan",
""
],
[
"Liu",
"Deguang",
""
],
[
"Li",
"Xin",
""
],
[
"Chen",
"Enhong",
""
]
] |
2506.11649
|
Matt Zepf
|
Michal Sm\'id, Pooyan Khademi, Carsten B\"ahtz, Erik Brambrink, Jindrich Chalupsky, Tom E. Cowan, Samuele Di Dio Cafiso, Sebastian G\"ode, J\"org Grenzer, Vera Hajkova, Peter Hilz, Willi Hippler, Hauke H\"opner, Alzbeta Horynova, Oliver Humphries, Simon Jelinek, Libor Juha, Felix Karbstein, Alejandro Laso-Garcia, Robert L\"otzsch, Aim\'e Math\'eron, Gerhard G. Paulus, Lisa Randolph, Alexander S\"avert, Hans-Peter Schlenvoigt, Jan Patrick Schwinekendorf, Thomas St\"ohlker, Toma Toncian, Maxim Valialshchikov, Edgar Weckert, Colin Wessel and Matt Zepf
|
The Darkfield Approach to Measuring Vacuum Birefringence and Light-by-Light Couplings -- A Proof-of-Principle Experiment
| null |
Phys.Rev.A 112 (2025) 6, 063512
|
10.1103/xpxy-ntwz
|
V11a
|
hep-ex physics.optics quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Vacuum fluctuations give rise to effective nonlinear interactions between electromagnetic fields. These generically modify the characteristics of light traversing a strong-field region. X-ray free-electron lasers constitute a particularly promising probe, due to their brilliance, the possibility of precise control and favourable frequency scaling. However, the nonlinear vacuum response is very small even when probing a tightly focused high-intensity laser field with XFEL radiation and direct measurement of light-by-light scattering of real photons and the associated fundamental physics constants of the quantum vacuum has not been possible to date. Achieving a sufficiently good signal-to-background separation is key to a successful quantum vacuum experiment. To master this challenge, a darkfield detection concept has recently been proposed. Here we present the results of a proof-of-principle experiment validating this approach at the High Energy Density scientific instrument of the European X-Ray Free Electron Laser.
|
[
{
"version": "v1",
"created": "Fri, 13 Jun 2025 10:23:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Smíd",
"Michal",
""
],
[
"Khademi",
"Pooyan",
""
],
[
"Bähtz",
"Carsten",
""
],
[
"Brambrink",
"Erik",
""
],
[
"Chalupsky",
"Jindrich",
""
],
[
"Cowan",
"Tom E.",
""
],
[
"Cafiso",
"Samuele Di Dio",
""
],
[
"Göde",
"Sebastian",
""
],
[
"Grenzer",
"Jörg",
""
],
[
"Hajkova",
"Vera",
""
],
[
"Hilz",
"Peter",
""
],
[
"Hippler",
"Willi",
""
],
[
"Höpner",
"Hauke",
""
],
[
"Horynova",
"Alzbeta",
""
],
[
"Humphries",
"Oliver",
""
],
[
"Jelinek",
"Simon",
""
],
[
"Juha",
"Libor",
""
],
[
"Karbstein",
"Felix",
""
],
[
"Laso-Garcia",
"Alejandro",
""
],
[
"Lötzsch",
"Robert",
""
],
[
"Mathéron",
"Aimé",
""
],
[
"Paulus",
"Gerhard G.",
""
],
[
"Randolph",
"Lisa",
""
],
[
"Sävert",
"Alexander",
""
],
[
"Schlenvoigt",
"Hans-Peter",
""
],
[
"Schwinekendorf",
"Jan Patrick",
""
],
[
"Stöhlker",
"Thomas",
""
],
[
"Toncian",
"Toma",
""
],
[
"Valialshchikov",
"Maxim",
""
],
[
"Weckert",
"Edgar",
""
],
[
"Wessel",
"Colin",
""
],
[
"Zepf",
"Matt",
""
]
] |
2506.12235
|
Leandro Roza Livramento PhD
|
L. A. Ferreira and L. R. Livramento
|
A Generalized False Vacuum Skyrme model
|
38 pages, 10 figures, 3 tables
| null | null | null |
hep-th math-ph math.MP nlin.SI nucl-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We propose a generalization of the False Vacuum Skyrme model for any simple compact Lie groups $G$ that leads to Hermitian symmetric spaces. The Skyrme field that in the original theory takes its values in $SU(2)$ Lie group, now takes its values in $G$, while the remaining fields correspond to the entries of a symmetric, positive, and invertible $\dim G \times \dim G$-dimensional matrix $h$. This model is also an extension of the generalized BPS Skyrme model. We prove that the global minima correspond to the $h$ fields being self-dual solutions of the generalized BPS Skyrme model, together with a particular field configuration for the Skyrme field that leads to a spherically symmetric topological charge density. As in the case of the original model, the minimization of the energy leads to the so-called reduced problem, defined in the context of false vacuum decay. This imposes a condition on the Skyrme field, which, if satisfied, makes the total energy of the global minima, as well as the main properties of the model, equivalent to those obtained for the $G=SU(2)$ case. We study this condition and its consequences within the generalized rational map ansatz and show how it can be satisfied for $G=SU(p+q)$, where $p$ and $q$ are positive integers, with the Hermitian symmetric spaces being $SU(p+q)/SU(p) \otimes SU(q) \otimes U(1)$. In such a case, the model is completely equivalent to the $G=SU(2)$ False Vacuum Skyrme model, independent of the values of $p$ and $q$. We also provide a numerical study of the baryon density, RMS radius, and binding energy per nucleon, which deepens the previous analysis conducted for the $SU(2)$ False Vacuum Skyrme model. Additionaly, in the case of $G = SU(3)$, we have studied the application of our model to the description of the binding energies and masses of the $\Lambda$-hypernuclei.
|
[
{
"version": "v1",
"created": "Fri, 13 Jun 2025 21:25:25 GMT"
},
{
"version": "v2",
"created": "Fri, 04 Jul 2025 23:00:44 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 12:29:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Ferreira",
"L. A.",
""
],
[
"Livramento",
"L. R.",
""
]
] |
2506.12305
|
Weiyang Wang
|
Zi-Hao Xu, Wei-Yang Wang, Shun-Shun Cao, Ren-Xin Xu
|
Pulsar Sparking: What if mountains on the surface?
|
15 pages, 6 figures, RAA accepted
| null | null | null |
astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
A numerical framework to calculate the height and potential of the vacuum inner gap of pulsars is presented here. %
The results demonstrate that small mountains on a pulsar's polar cap tend to significantly influence the properties of the inner vacuum gap, making it easier for sparks to form. %
In this scenario, the magnetospheric activity observed from the pulsars PSR J0250$+$5854 and PSR J2144$-$3933 which lie below the traditional pulsar death line, and some single-pulse modulation phenomena could also then be understood. %
Furthermore, the presence of small mountains should depend on the puzzling state of supranuclear matter inside pulsars. %
In order to sustain stable mountains on the surface, pulsars might be made of solid strangeon matter, which is favoured by both the charge neutrality and the flavour symmetry of quarks.
|
[
{
"version": "v1",
"created": "Sat, 14 Jun 2025 01:46:29 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 10:26:03 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Zi-Hao",
""
],
[
"Wang",
"Wei-Yang",
""
],
[
"Cao",
"Shun-Shun",
""
],
[
"Xu",
"Ren-Xin",
""
]
] |
2506.13018
|
Bo Zhao
|
Bo Zhao, Robin Walters, Rose Yu
|
Symmetry in Neural Network Parameter Spaces
|
Published in Transactions on Machine Learning Research (TMLR)
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Modern deep learning models are highly overparameterized, resulting in large sets of parameter configurations that yield the same outputs. A significant portion of this redundancy is explained by symmetries in the parameter space--transformations that leave the network function unchanged. These symmetries shape the loss landscape and constrain learning dynamics, offering a new lens for understanding optimization, generalization, and model complexity that complements existing theory of deep learning. This survey provides an overview of parameter space symmetry. We summarize existing literature, uncover connections between symmetry and learning theory, and identify gaps and opportunities in this emerging field.
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 00:59:12 GMT"
},
{
"version": "v2",
"created": "Fri, 10 Oct 2025 04:33:49 GMT"
},
{
"version": "v3",
"created": "Wed, 10 Dec 2025 19:34:39 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhao",
"Bo",
""
],
[
"Walters",
"Robin",
""
],
[
"Yu",
"Rose",
""
]
] |
2506.13270
|
Nan Chen
|
Nan Chen, Luna K. Qiu, Arran Zeyu Wang, Zilong Wang, Yuqing Yang
|
Screen Reader Programmers in the Vibe Coding Era: Adaptation, Empowerment, and New Accessibility Landscape
| null | null | null | null |
cs.HC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Generative AI agents are reshaping human-computer interaction, shifting users from direct task execution to supervising machine-driven actions, especially the rise of "vibe coding" in programming. Yet little is known about how screen reader programmers interact with AI code assistants in practice. We conducted a longitudinal study with 16 blind and low-vision programmers. Participants completed a GitHub Copilot tutorial, engaged with a programming task, and provided initial feedback. After two weeks of AI-assisted programming, follow-ups examined how their practices and perceptions evolved. Our findings show that code assistants enhanced programming efficiency and bridged accessibility gaps. However, participants struggled to convey intent, interpret AI outputs, and manage multiple views while maintaining situational awareness. They showed diverse preferences for accessibility features, expressed a need to balance automation with control, and encountered barriers when learning to use these tools. Furthermore, we propose design principles and recommendations for more accessible and inclusive human-AI collaborations.
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 09:11:02 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 05:33:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chen",
"Nan",
""
],
[
"Qiu",
"Luna K.",
""
],
[
"Wang",
"Arran Zeyu",
""
],
[
"Wang",
"Zilong",
""
],
[
"Yang",
"Yuqing",
""
]
] |
2506.13864
|
Alessandro Lenoci
|
Francesco D'Eramo, Alessandro Lenoci and Ariane Dekker
|
Dark Matter Freeze-In and Small-Scale Observables: Novel Mass Bounds and Viable Particle Candidates
|
17 pages, 7 figures, 2 tables, version published in PRD
|
Phys.Rev.D 112 (2025) 11, 116008
|
10.1103/j62q-cvkr
| null |
hep-ph astro-ph.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The suppression of cosmological structure at small scales is a key signature of dark matter (DM) produced via freeze-in in the low-mass regime. We present a comprehensive analysis of its impact, incorporating recent constraints from Milky Way satellite counts, strong gravitational lensing with JWST data, and the Lyman-$\alpha$ forest. We adopt a general strategy to translate existing warm dark matter (WDM) bounds into lower mass limits for a broad class of DM candidates characterized by quasi-thermal phase space distributions. The benefits of this approach include computational efficiency and the ability to explore a wide range of models. We derive model-independent bounds for DM produced via two-body decays, scatterings, and three-body decays, and apply the framework to concrete scenarios such as the Higgs portal, sterile neutrinos, axion-like particles, and the dark photon portal. Results from specific models confirm the validity of the model-independent analysis.
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 18:00:02 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 14:56:04 GMT"
},
{
"version": "v3",
"created": "Wed, 10 Dec 2025 08:02:59 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"D'Eramo",
"Francesco",
""
],
[
"Lenoci",
"Alessandro",
""
],
[
"Dekker",
"Ariane",
""
]
] |
2506.14022
|
Jacob Landsberg
|
Jacob B. Landsberg, Elizabeth A. Barnes, Matthew Newman
|
AI-Informed Model Analogs for Subseasonal-to-Seasonal Prediction
|
23 pages, 12 figures
| null | null | null |
physics.ao-ph cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Subseasonal-to-seasonal forecasting is crucial for public health, disaster preparedness, and agriculture, and yet it remains a particularly challenging timescale to predict. We explore the use of an interpretable AI-informed model analog forecasting approach, previously employed on longer timescales, to improve S2S predictions. Using an artificial neural network, we learn a mask of weights to optimize analog selection and showcase its versatility across three varied prediction tasks: 1) classification of Week 3-4 Southern California summer temperatures; 2) regional regression of Month 1 midwestern U.S. summer temperatures; and 3) classification of Month 1-2 North Atlantic wintertime upper atmospheric winds. The AI-informed analogs outperform traditional analog forecasting approaches, as well as climatology and persistence baselines, for deterministic and probabilistic skill metrics on both climate model and reanalysis data. We find the analog ensembles built using the AI-informed approach also produce better predictions of temperature extremes and improve representation of forecast uncertainty. Finally, by using an interpretable-AI framework, we analyze the learned masks of weights to better understand S2S sources of predictability.
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 21:39:13 GMT"
},
{
"version": "v2",
"created": "Wed, 10 Dec 2025 21:08:24 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Landsberg",
"Jacob B.",
""
],
[
"Barnes",
"Elizabeth A.",
""
],
[
"Newman",
"Matthew",
""
]
] |
2506.14140
|
Guo-Li Wang
|
Shi-Hang Zhang, Wen-Yuan Ke, Su-Yan Pei, Wei Li, Xiao-Ze Tan, Lili Zhu, Guo-Li Wang
|
Identification of $D^*_2(3000)$ as the $D_2^*(2^3P_2)$ and exploring potential of undiscovered $2^+$ mesons via $B$ decays
|
28 pages, 7 figures
|
Phys.Rev.D 112 (2025), 113001
| null | null |
hep-ph hep-ex
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Following the discovery of the $D^*_2(3000)$, its mass and full width have been extensively studied. Yet its nature remains undetermined to date. Since it was discovered through nonleptonic decay of $B$ meson and the corresponding cascade process, we therefore in this paper investigate the nonleptonic and semileptonic decays of $B$ meson to $J^P = 2^+$ charmed mesons using the Bethe-Salpeter equation approach. Our calculations on nonleptonic $B$ decays reveal that the unconfirmed resonance $D^*_2(3000)$ aligns well with $D^*_2(2^3P_2)$ predictions. Other candidates, including $D^*_2(1^3F_2)$, $D^*_2(3^3P_2)$, and $D^*_2(2^3F_2)$, are excluded due to their very small branching ratios in $B$ decays. Considering that the $D^*_2(1F)$, $D^*_2(3P)$, and $D^*_2(2F)$ have not yet been experimentally observed, we investigate the feasibility of their detection in $B$-meson decays.
|
[
{
"version": "v1",
"created": "Tue, 17 Jun 2025 03:05:59 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Nov 2025 11:09:05 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 08:11:00 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Zhang",
"Shi-Hang",
""
],
[
"Ke",
"Wen-Yuan",
""
],
[
"Pei",
"Su-Yan",
""
],
[
"Li",
"Wei",
""
],
[
"Tan",
"Xiao-Ze",
""
],
[
"Zhu",
"Lili",
""
],
[
"Wang",
"Guo-Li",
""
]
] |
2506.14559
|
Noah Grodzinski
|
Noah Grodzinski, Michael E. Cates, Robert L. Jack
|
Hydrodynamic theory of wetting by active particles
|
16 pages, 7 figures
| null | null | null |
cond-mat.stat-mech cond-mat.soft
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The accumulation of self-propelled particles on repulsive barriers is a widely observed feature in active matter. Despite being implicated in a broad range of biological processes, from biofilm formation to cytoskeletal movement, wetting of surfaces by active particles remains poorly understood. In this work, we study this active wetting by considering a model comprising an active lattice gas, interacting with a permeable barrier under periodic boundary conditions, for which an exact hydrodynamic description is possible. We consider a hydrodynamic scaling limit that eliminates dynamical noise while retaining microscopic interpretability, enabling a precise characterisation of steady-states and their transitions. We demonstrate that the accumulation of active particles has remarkable similarities to equilibrium wetting, and that active wetting transitions display all the salient characteristics of the equilibrium critical wetting transition -- despite fundamental differences in underlying microscopic dynamics. However, our framework also enables the investigation of subtle but important nonequilibrium effects in active wetting, including a spontaneous ratchet effect which leads to a global steady-state current, departure of bulk densities from their binodal values, and a novel dynamical transition pathway. Our results provide an intrinsically nonequilibrium framework in which to study active wetting, precisely demonstrating the connection to passive wetting while clarifying the nonequilibrium consequences of activity.
|
[
{
"version": "v1",
"created": "Tue, 17 Jun 2025 14:14:51 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 16:44:26 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Grodzinski",
"Noah",
""
],
[
"Cates",
"Michael E.",
""
],
[
"Jack",
"Robert L.",
""
]
] |
2506.14740
|
Chin-Wen Chou
|
Dalton Chaffee, Baruch Margulis, April Sheffield, Julian Schmidt, April Reisenfeld, David R. Leibrandt, Dietrich Leibfried, Chin-Wen Chou
|
High-fidelity quantum state control of a polar molecular ion in a cryogenic environment
|
Updated references and supplementary information
|
Phys. Rev. Lett. 135, 240801 (2025)
|
10.1103/7ypf-91jr
| null |
physics.atom-ph quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We use a quantum-logic spectroscopy (QLS) protocol to control the quantum state of a CaH+ ion in a cryogenic environment, in which reduced thermal radiation extends rotational state lifetimes by an order of magnitude over those at room temperature. By repeatedly and adaptively probing the molecule, detecting the outcome of each probe via an atomic ion, and using a Bayesian update scheme to quantify confidence in the molecular state, we demonstrate state preparation and measurement (SPAM) in a single quantum state with infidelity less than 6x10^-3 and measure Rabi flopping between two states with greater than 99% contrast. The protocol does not require any molecule-specific lasers and the detection scheme is non-destructive.
|
[
{
"version": "v1",
"created": "Tue, 17 Jun 2025 17:20:10 GMT"
},
{
"version": "v2",
"created": "Sat, 19 Jul 2025 01:52:34 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Chaffee",
"Dalton",
""
],
[
"Margulis",
"Baruch",
""
],
[
"Sheffield",
"April",
""
],
[
"Schmidt",
"Julian",
""
],
[
"Reisenfeld",
"April",
""
],
[
"Leibrandt",
"David R.",
""
],
[
"Leibfried",
"Dietrich",
""
],
[
"Chou",
"Chin-Wen",
""
]
] |
2506.14933
|
Adriana Watson
|
Adriana Watson, Grant Richards, Daniel Schiff
|
Explain First, Trust Later: LLM-Augmented Explanations for Graph-Based Crypto Anomaly Detection
|
11 pages, 5 figures. Code available at: https://github.com/awatson246/crypto-anomaly-detection-policy
| null | null | null |
cs.CE cs.AI cs.CR
|
http://creativecommons.org/licenses/by/4.0/
|
The decentralized finance (DeFi) community has grown rapidly in recent years, pushed forward by cryptocurrency enthusiasts interested in the vast untapped potential of new markets. The surge in popularity of cryptocurrency has ushered in a new era of financial crime. Unfortunately, the novelty of the technology makes the task of catching and prosecuting offenders particularly challenging. Thus, it is necessary to implement automated detection tools related to policies to address the growing criminality in the cryptocurrency realm.
|
[
{
"version": "v1",
"created": "Tue, 17 Jun 2025 19:30:21 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 03:15:22 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Watson",
"Adriana",
""
],
[
"Richards",
"Grant",
""
],
[
"Schiff",
"Daniel",
""
]
] |
2506.15881
|
Alexey Yermakov
|
Alexey Yermakov, David Zoro, Mars Liyao Gao and J. Nathan Kutz
|
T-SHRED: Symbolic Regression for Regularization and Model Discovery with Transformer Shallow Recurrent Decoders
|
17 pages, 5 figures, submitted to Transactions of the Royal Society (Symbolic Regression in the Physical Sciences)
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
SHallow REcurrent Decoders (SHRED) are effective for system identification and forecasting from sparse sensor measurements. Such models are light-weight and computationally efficient, allowing them to be trained on consumer laptops. SHRED-based models rely on Recurrent Neural Networks (RNNs) and a simple Multi-Layer Perceptron (MLP) for the temporal encoding and spatial decoding respectively. Despite the relatively simple structure of SHRED, they are able to predict chaotic dynamical systems on different physical, spatial, and temporal scales directly from a sparse set of sensor measurements. In this work, we modify SHRED by leveraging transformers (T-SHRED) embedded with symbolic regression for the temporal encoding, circumventing auto-regressive long-term forecasting for physical data. This is achieved through a new sparse identification of nonlinear dynamics (SINDy) attention mechanism into T-SHRED to impose sparsity regularization on the latent space, which also allows for immediate symbolic interpretation. Symbolic regression improves model interpretability by learning and regularizing the dynamics of the latent space during training. We analyze the performance of T-SHRED on three different dynamical systems ranging from low-data to high-data regimes.
|
[
{
"version": "v1",
"created": "Wed, 18 Jun 2025 21:14:38 GMT"
},
{
"version": "v2",
"created": "Fri, 21 Nov 2025 19:50:20 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 17:28:30 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Yermakov",
"Alexey",
""
],
[
"Zoro",
"David",
""
],
[
"Gao",
"Mars Liyao",
""
],
[
"Kutz",
"J. Nathan",
""
]
] |
2506.16528
|
Bornali Phukon
|
Bornali Phukon, Xiuwen Zheng, Mark Hasegawa-Johnson
|
Aligning ASR Evaluation with Human and LLM Judgments: Intelligibility Metrics Using Phonetic, Semantic, and NLI Approaches
|
5 pages, 2 figures, Interspeech 2025
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Traditional ASR metrics like WER and CER fail to capture intelligibility, especially for dysarthric and dysphonic speech, where semantic alignment matters more than exact word matches. ASR systems struggle with these speech types, often producing errors like phoneme repetitions and imprecise consonants, yet the meaning remains clear to human listeners. We identify two key challenges: (1) Existing metrics do not adequately reflect intelligibility, and (2) while LLMs can refine ASR output, their effectiveness in correcting ASR transcripts of dysarthric speech remains underexplored. To address this, we propose a novel metric integrating Natural Language Inference (NLI) scores, semantic similarity, and phonetic similarity. Our ASR evaluation metric achieves a 0.890 correlation with human judgments on Speech Accessibility Project data, surpassing traditional methods and emphasizing the need to prioritize intelligibility over error-based measures.
|
[
{
"version": "v1",
"created": "Thu, 19 Jun 2025 18:21:19 GMT"
},
{
"version": "v2",
"created": "Thu, 11 Dec 2025 07:36:11 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Phukon",
"Bornali",
""
],
[
"Zheng",
"Xiuwen",
""
],
[
"Hasegawa-Johnson",
"Mark",
""
]
] |
2506.16685
|
Xiaomeng Xu
|
Xiaomeng Xu, Yifan Hou, Zeyi Liu, Shuran Song
|
Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Corrections
| null | null | null | null |
cs.RO cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
We address key challenges in Dataset Aggregation (DAgger) for real-world contact-rich manipulation: how to collect informative human correction data and how to effectively update policies with this new data. We introduce Compliant Residual DAgger (CR-DAgger), which contains two novel components: 1) a Compliant Intervention Interface that leverages compliance control, allowing humans to provide gentle, accurate delta action corrections without interrupting the ongoing robot policy execution; and 2) a Compliant Residual Policy formulation that learns from human corrections while incorporating force feedback and force control. Our system significantly enhances performance on precise contact-rich manipulation tasks using minimal correction data, improving base policy success rates by over 50\% on two challenging tasks (book flipping and belt assembly) while outperforming both retraining-from-scratch and finetuning approaches. Through extensive real-world experiments, we provide practical guidance for implementing effective DAgger in real-world robot learning tasks. Result videos are available at: https://compliant-residual-dagger.github.io/
|
[
{
"version": "v1",
"created": "Fri, 20 Jun 2025 01:57:47 GMT"
},
{
"version": "v2",
"created": "Thu, 24 Jul 2025 04:54:15 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Dec 2025 10:37:47 GMT"
},
{
"version": "v4",
"created": "Thu, 11 Dec 2025 10:15:43 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Xu",
"Xiaomeng",
""
],
[
"Hou",
"Yifan",
""
],
[
"Liu",
"Zeyi",
""
],
[
"Song",
"Shuran",
""
]
] |
2506.17090
|
Matthew Finlayson
|
Murtaza Nazir, Matthew Finlayson, John X. Morris, Xiang Ren, Swabha Swayamdipta
|
Better Language Model Inversion by Compactly Representing Next-Token Distributions
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Language model inversion seeks to recover hidden prompts using only language model outputs. This capability has implications for security and accountability in language model deployments, such as leaking private information from an API-protected language model's system message. We propose a new method -- prompt inversion from logprob sequences (PILS) -- that recovers hidden prompts by gleaning clues from the model's next-token probabilities over the course of multiple generation steps. Our method is enabled by a key insight: The vector-valued outputs of a language model occupy a low-dimensional subspace. This enables us to losslessly compress the full next-token probability distribution over multiple generation steps using a linear map, allowing more output information to be used for inversion. Our approach yields massive gains over previous state-of-the-art methods for recovering hidden prompts, achieving 2--3.5 times higher exact recovery rates across test sets, in one case increasing the recovery rate from 17% to 60%. Our method also exhibits surprisingly good generalization behavior; for instance, an inverter trained on 16 generations steps gets 5--27 points higher prompt recovery when we increase the number of steps to 32 at test time. Furthermore, we demonstrate strong performance of our method on the more challenging task of recovering hidden system messages. We also analyze the role of verbatim repetition in prompt recovery and propose a new method for cross-family model transfer for logit-based inverters. Our findings show that next-token probabilities are a considerably more vulnerable attack surface for inversion attacks than previously known.
|
[
{
"version": "v1",
"created": "Fri, 20 Jun 2025 15:53:51 GMT"
},
{
"version": "v2",
"created": "Mon, 23 Jun 2025 14:39:37 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 17:53:55 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Nazir",
"Murtaza",
""
],
[
"Finlayson",
"Matthew",
""
],
[
"Morris",
"John X.",
""
],
[
"Ren",
"Xiang",
""
],
[
"Swayamdipta",
"Swabha",
""
]
] |
2506.17246
|
Samad Hajinazar
|
Samad Hajinazar and Eva Zurek
|
XtalOpt Version 14: Variable-Composition Crystal Structure Search for Functional Materials Through Pareto Optimization
| null | null | null | null |
physics.comp-ph cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
Version 14 of XtalOpt, an evolutionary multi-objective global optimization algorithm for crystal structure prediction, is now available for download from its official website https://xtalopt.github.io, and the Computer Physics Communications Library. The new version of the code is designed to perform a ground state search for crystal structures with variable compositions by integrating a suite of ab initio methods alongside classical and machine-learning potentials for structural relaxation. The multi-objective search framework has been enhanced through the introduction of Pareto optimization, enabling efficient discovery of functional materials. Herein, we describe the newly implemented methodologies, provide detailed instructions for their use, and present an overview of additional improvements included in the latest version of the code.
|
[
{
"version": "v1",
"created": "Sat, 07 Jun 2025 04:59:48 GMT"
},
{
"version": "v2",
"created": "Sun, 20 Jul 2025 03:06:33 GMT"
},
{
"version": "v3",
"created": "Thu, 11 Dec 2025 01:06:15 GMT"
}
] | 2025-12-12T00:00:00
|
[
[
"Hajinazar",
"Samad",
""
],
[
"Zurek",
"Eva",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.