3 papers
cs.AI2026
DASH: Divergence-Adaptive Supervision Horizons for On-Policy Self-Distillation of Reasoning Models
ZhiYan Hou, Xinyu Tang, Hongyan An +9
Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models using automatically verifiable outcome signals, but these signals…
cs.RO2026
Minimizing Worst-Case Weighted Latency for Multi-Robot Persistent Monitoring: Theory and RL-Based Solutions
Weizhen Wang, Ziheng Wang, Jianping He +2
We study multi-robot persistent monitoring on weighted graphs, where node weights encode monitoring priorities and edge weights encode travel distances. The goal is to design joint…
cs.LG2025
Analysis of On-policy Policy Gradient Methods under the Distribution Mismatch
Weizhen Wang, Jianping He, Xinzhe Zhou +1
Policy gradient methods are among the most successful approaches for solving challenging reinforcement learning problems. Despite their empirical success, many state-of-the-art pol…