6 papers
FAIR-Calib: Frontier-Aware Instability-Reweighted Calibration for Post-Training Quantization of Diffusion Large Language Models
Haoyu Huang, Linlin Yang, Sheng Xu +5
Diffusion Large Language Models (dLLMs) refine tokens iteratively but commit them irreversibly, leading to a "stability lag" where early decisions remain fragile even after being w…
Where Hindsight Credit Can Reside: A Signed-Capacity View of Token Updates in RLVR
Yuhang He, Haodong Wu, Siyi Liu +7
Reinforcement Learning with Verifiable Rewards (RLVR) improves the reasoning ability of Large Language Models (LLMs), but sparse outcome rewards make token-level credit assignment…
GRAM-R: Self-Training Generative Foundation Reward Models for Reward Reasoning
Chenglong Wang, Yongyu Mu, Hang Zhou +10
Significant progress in reward modeling over recent years has been driven by a paradigm shift from task-specific designs towards generalist reward models. Despite this trend, devel…
MRO: Enhancing Reasoning in Diffusion Language Models via Multi-Reward Optimization
Chenglong Wang, Yang Gan, Hang Zhou +10
Recent advances in diffusion language models (DLMs) have presented a promising alternative to traditional autoregressive large language models (LLMs). However, DLMs still lag behin…
Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models
Kaiyan Chang, Yonghao Shi, Chenglong Wang +7
Test-Time Scaling (TTS) is a promising approach to progressively elicit the model's intelligence during inference. Recently, training-based TTS methods, such as continued reinforce…
Learning Evaluation Models from Large Language Models for Sequence Generation
Chenglong Wang, Hang Zhou, Kaiyan Chang +6
Automatic evaluation of sequence generation, traditionally reliant on metrics like BLEU and ROUGE, often fails to capture the semantic accuracy of generated text sequences due to t…