debaterhub/sentence-selection-orpo-lora

Sentence Selection ORPO LoRA

Fine-tuned LoRA adapter for sentence selection in debate contexts.

Training Details

• Base Model: Qwen3-30B-A3B (via SFT fine-tuned version)
• Method: ORPO (Odds Ratio Preference Optimization)
• Task: Select relevant sentence IDs from academic texts to support claims
• F1 Score: 0.247 on holdout set

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM

base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-30B-A3B")
model = PeftModel.from_pretrained(base_model, "debaterhub/sentence-selection-orpo-lora")

Key Findings

1. Format-consistent DPO pairs essential for ORPO training
2. 2 epochs optimal (more causes overfitting)
3. Noise augmentation fixes positional bias in training data