14 papers
QUADS: Stabilizing NVFP4 Reinforcement Learning for MoE via QUantization-error Alignment across Dual Sides
Zhengyang Zhuge, Hao Yu, Xin Wang +4
Rollout generation is a major bottleneck in Reinforcement Learning (RL) for Mixture-of-Experts (MoE) Large Language Models, motivating low-precision rollout acceleration such as FP…
Survive or Collapse: The Asymmetric Roles of Data Gating and Reward Grounding in Self-Play RL
Sophia Xiao Pu, Zhaotian Weng, Chengzhi Liu +4
Self-play reinforcement learning trains language models on their own generated tasks, co-evolving a proposer and solver without human labels. Recent systems report strong reasoning…
ECPO: Evidence-Coupled Policy Optimization for Evidence-Certified Candidate Ranking
Miaobo Hu, Shuhao Hu, BoKun Wang +5
Ranking systems used in decision-support settings should not only order candidates but also expose evidence that can be independently checked. We study evidence-certified candidate…
Towards Context-Invariant Safety Alignment for Large Language Models
Yixu Wang, Yang Yao, Xin Wang +4
Preference-based post-training aligns LLMs with human intent, yet safety behavior often remains brittle. A model may refuse a harmful request in a standard prompt but comply when t…
AGPO: Adaptive Group Policy Optimization with Dual Statistical Feedback
Miaobo Hu, Shuhao Hu, Bokun Wang +5
Reinforcement learning improves LLM reasoning, but PPO/GRPO typically use fixed clipping and decoding temperature, which makes training brittle and tuning-heavy. We propose Adaptiv…
SAVER: Selective As-Needed Vision Evidence for Multimodal Information Extraction
Miaobo Hu, Shuhao Hu, Bokun Wang +5
Multimodal IE in social media is difficult because a post may attach multiple images that are weakly related, redundant, or even misleading with respect to the text. In this settin…