6 papers · 1 filter
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…
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…
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…
Hybrid Alignment Training for Large Language Models
Chenglong Wang, Hang Zhou, Kaiyan Chang +5
Alignment training is crucial for enabling large language models (LLMs) to cater to human intentions and preferences. It is typically performed based on two stages with different o…
Prior Constraints-based Reward Model Training for Aligning Large Language Models
Hang Zhou, Chenglong Wang, Yimin Hu +3
Reinforcement learning with human feedback for aligning large language models (LLMs) trains a reward model typically using ranking loss with comparison pairs.However, the training…
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…