TreeRL: LLM Reinforcement Learning with On-Policy Tree Search

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Hou, Zhenyu, Hu, Ziniu, Li, Yujiang, Lu, Rui, Tang, Jie, Dong, Yuxiao
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

_version_	1866916793453379584
author	Hou, Zhenyu Hu, Ziniu Li, Yujiang Lu, Rui Tang, Jie Dong, Yuxiao
author_facet	Hou, Zhenyu Hu, Ziniu Li, Yujiang Lu, Rui Tang, Jie Dong, Yuxiao
contents	Reinforcement learning (RL) with tree search has demonstrated superior performance in traditional reasoning tasks. Compared to conventional independent chain sampling strategies with outcome supervision, tree search enables better exploration of the reasoning space and provides dense, on-policy process rewards during RL training but remains under-explored in On-Policy LLM RL. We propose TreeRL, a reinforcement learning framework that directly incorporates on-policy tree search for RL training. Our approach includes intermediate supervision and eliminates the need for a separate reward model training. Existing approaches typically train a separate process reward model, which can suffer from distribution mismatch and reward hacking. We also introduce a cost-effective tree search approach that achieves higher search efficiency under the same generation token budget by strategically branching from high-uncertainty intermediate steps rather than using random branching. Experiments on challenging math and code reasoning benchmarks demonstrate that TreeRL achieves superior performance compared to traditional ChainRL, highlighting the potential of tree search for LLM. TreeRL is open-sourced at https://github.com/THUDM/TreeRL.
format	Preprint
id	arxiv_https___arxiv_org_abs_2506_11902
institution	arXiv
publishDate	2025
record_format	arxiv
spellingShingle	TreeRL: LLM Reinforcement Learning with On-Policy Tree Search Hou, Zhenyu Hu, Ziniu Li, Yujiang Lu, Rui Tang, Jie Dong, Yuxiao Machine Learning Computation and Language Reinforcement learning (RL) with tree search has demonstrated superior performance in traditional reasoning tasks. Compared to conventional independent chain sampling strategies with outcome supervision, tree search enables better exploration of the reasoning space and provides dense, on-policy process rewards during RL training but remains under-explored in On-Policy LLM RL. We propose TreeRL, a reinforcement learning framework that directly incorporates on-policy tree search for RL training. Our approach includes intermediate supervision and eliminates the need for a separate reward model training. Existing approaches typically train a separate process reward model, which can suffer from distribution mismatch and reward hacking. We also introduce a cost-effective tree search approach that achieves higher search efficiency under the same generation token budget by strategically branching from high-uncertainty intermediate steps rather than using random branching. Experiments on challenging math and code reasoning benchmarks demonstrate that TreeRL achieves superior performance compared to traditional ChainRL, highlighting the potential of tree search for LLM. TreeRL is open-sourced at https://github.com/THUDM/TreeRL.
title	TreeRL: LLM Reinforcement Learning with On-Policy Tree Search
topic	Machine Learning Computation and Language
url	https://arxiv.org/abs/2506.11902