11 papers
PolyAlign: Conditional Human-Distribution Alignment
L. D. M. S. Sai Teja, Ufaq Khan, Sathira Silva +2
Post-training methods such as supervised fine-tuning (SFT) and preference optimization typically align language models toward a single global assistant behavior. While effective fo…
RLBFF: Binary Flexible Feedback to bridge between Human Feedback & Verifiable Rewards
Zhilin Wang, Jiaqi Zeng, Olivier Delalleau +6
Reinforcement Learning with Human Feedback (RLHF) and Reinforcement Learning with Verifiable Rewards (RLVR) are the main RL paradigms used in LLM post-training, each offering disti…
Diverging Preferences: When do Annotators Disagree and do Models Know?
Michael JQ Zhang, Zhilin Wang, Jena D. Hwang +6
We examine diverging preferences in human-labeled preference datasets. We develop a taxonomy of disagreement sources spanning ten categories across four high-level classes and find…
PhyCritic: Multimodal Critic Models for Physical AI
Tianyi Xiong, Shihao Wang, Guilin Liu +5
With the rapid development of large multimodal models, reliable judge and critic models have become essential for open-ended evaluation and preference alignment, providing pairwise…
REF-VLM: Triplet-Based Referring Paradigm for Unified Visual Decoding
Yan Tai, Luhao Zhu, Yunan Ding +4
Multimodal Large Language Models (MLLMs) demonstrate robust zero-shot capabilities across diverse vision-language tasks after training on mega-scale datasets. However, dense predic…
HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languages
Zhilin Wang, Jiaqi Zeng, Olivier Delalleau +6
Preference datasets are essential for training general-domain, instruction-following language models with Reinforcement Learning from Human Feedback (RLHF). Each subsequent data re…