arXiv AI By Yuyao Zheng, Haipeng Sun, Junwei Bao, Lemao Liu, Hongfei Jiang, Yang Song, Dejing Dou

PGPO: Potential-Guided Policy Optimization for Multi-Turn Agentic Tasks

Read the original on arXiv AI →

The paper introduces Potential-Guided Policy Optimization (PGPO), a method for multi-turn agentic tasks that improves credit assignment by estimating empirical state potentials from anchor-state-group return statistics. PGPO derives action advantages from potential differences between adjacent states, enabling cross-trajectory credit propagation and finer-grained step-level credit assignment, especially within failed trajectories. Experiments on ALFWorld and WebShop demonstrate strong performance compared to recent group-based reinforcement learning methods, with negligible training overhead.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Hugging Face Trending Papers
Sep 2

PGPO: Potential-Guided Policy Optimization for Multi-Turn Agentic Tasks

PGPO: Potential-Guided Policy Optimization for Multi-Turn Agentic Tasks proposes a new reinforcement learning approach that estimates empirical state potentials from anchor-state-group return statistics within each rollout group. By deriving action advantages from potential differences between adjacent states, PGPO enables cross‑trajectory credit propagation, providing finer‑grained step‑level credit assignment, especially within failed trajectories. Experiments on ALFWorld and WebShop demonstrate strong overall performance compared to recent group‑based RL methods, with negligible training overhead.

arXiv AI
Jul 20

Process Reward Informed Tree Rollout for Effective Multi-Turn RL

arXiv:2607. 15610v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become a key approach for training LLM agents, yet popular methods such as GRPO/RLOO rely on multiple independently sampled complete trajectories for advantage estimation.

By Xintong Li, Sha Li, Yuwei Zhang, Changlong Yu, Rongmei Lin, Hongye Jin, Shuyi Guan, Xin Liu, Linwei Li, Qingyu Yin, Jingbo Shang
arXiv AI
Jun 2

Beyond Trajectory-Level Attribution: Graph-Based Credit Assignment for Agentic Reinforcement Learning

arXiv:2605. 26684v2 Announce Type: replace-cross Abstract: Group-based reinforcement learning (RL) methods have achieved remarkable success in improving the performance of large language models (LLMs) and have been rapidly extended to agentic tasks.

By Xin Cheng, Shuo He, Lang Feng, HaiYang Xu, Ming Yan, Lei Feng, Bo An