T2PO: Uncertainty-Guided Exploration Control for Stable Multi-Turn Agentic Reinforcement Learning

Haixin Wang, Hejie Cui, Chenwei Zhang, Xin Liu, Shuowei Jin, Shijie Geng, Xinyang Zhang, Nasser Zalmout, Zhenyu Shi, Yizhou Sun
May, 2026
PDF
Type
Conference paper
Publication
Forty-Third International Conference on Machine Learning
LLM Post-Training
Shuowei Jin
Shuowei Jin
Applied Scientist at Amazon

My research interests include efficient LLM inference/training algorithms/systems, LLM post-training recipe, and general machine learning systems.