A novel policy gradient method is introduced, Hindsight Policy Optimization (HPO), that projects both the current policy distribution and the hindsight distribution into an intent space and extracts low-variance learning signals from the Wasserstein distance between them.
Abstract
Reinforcement learning (RL) has become a widely adopted technique for improving large language models (LLMs) on complex tasks. Despite this progress, existing RL methods still face challenges in training agents with longer-horizon interactions. One major bottleneck is distinguishing the contribution of different actions in long-horizon interaction, leading to high optimization variance. To address this, we introduce a novel policy gradient method, Hindsight Policy Optimization (HPO), that projects both the current policy distribution and the hindsight distribution into an intent space and extracts low-variance learning signals from the Wasserstein distance between them. We theoretically and empirically show that aggregating semantically similar states and actions in the intent space yields a bounded-variance estimator and improves policy performance stably. Our code is available online.
Isolated Bilateral Reinforcement Learning (IB-RL), in which the two roles coevolve through joint rollouts while each role optimizes its own reward through fully independent advantages, action masks, and update paths, produces policies that generalize more effectively to unseen counterparts.
Senhao Wang, Chenghao Cai, Hai-Tao Hu et al.· 0 citations
This work formalizes the hybrid LLM-planner and RL-controller architecture as a Goal-Augmented Markov Decision Process and shows that when the LLM per-state progress score is used as a bounded potential function, the resulting shaping term preserves the optimal policy set even when the LLM scores are inaccurate.
Christophe D. Hounwanou, John Emeka Eze, Yaé Ulrich Gaba· 1 citation
Production large language model (LLM) based systems such as coding agents, web navigators, and tool-calling assistants operate over multiple turns of interaction with users, tools, and environments. Pretrained LLMs, depending on their size, can either underperform in these settings due to misalignment with the system's...
Zhe Wang, Sapana Chaudhary, Jiayu Li et al.· Proceedings of the 32nd ACM...· 0 citations
River, a simple training recipe that improves reward quality by filtering low-quality environments and augmenting outcome rewards with process-level behavior regularization is proposed, which achieves the best performance among evaluated open-source RL-trained 8B models across four terminal-agent benchmarks.
Yi-Fan Yao, Bo Pang, Xuan-Phi Nguyen et al.· 2 citations
Progress-conditioned Group Policy Optimization is proposed, which uses first-visit observation coverage only when all samples in a group receive zero outcome reward, and consistently improves over group-based baselines, with particularly large gains on hard tasks.
Kaibing Yang, Guangfeng Cai, Sheng-Tian Yang et al.· arXiv.org· 1 citation
This work proposes QWM, a framework that leverages world models to perform test-time search over imagined trajectories on top of Q-learning to select high-value actions during both online rollouts and evaluation, and significantly outperforms strong prior state-of-the-art methods on both sample efficiency and performan...
Perry Dong, Yue-Ru Jia, Chelsea Finn et al.· 2 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
