2 papers
cs.AI2026
Improving Math Reasoning through Value-guided Informative Search
Shaohuai Liu, Yuning Wu, Haoran Liu +3
Reinforcement learning with verifiable rewards (RLVR) has substantially improved the mathematical reasoning capabilities of large language models. Recent work introduces search int…
cs.LG2026
Physics-Guided Policy Optimization with Self-Distillation
Ke Wang, Yuning Wu, Haoran Liu +3
Self-distilled policy optimization (SDPO) has become a popular paradigm for LLM post-training, where a model learns from its own predictions conditioned on privileged information.…