Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

DPEPO: Diverse Parallel Exploration Policy Optimization for LLM-based Agents

About

Large language model (LLM) agents that follow the sequential "reason-then-act" paradigm have achieved superior performance in many complex tasks.However, these methods suffer from limited exploration and incomplete environmental understanding, as they interact with only a single environment per step. In this paper, we first introduce a novel paradigm that enables an agent to interact with multiple environments simultaneously and share cross-trajectory experiences. Building upon this paradigm, we further propose DPEPO, a reinforcement learning (RL) algorithm that encourages the agent to perform diverse parallel exploration. There are two stages in DPEPO: initial supervised fine-tuning (SFT) imparts basic parallel reasoning and action generation, followed by reinforcement learning stage with a hierarchical reward scheme. We design a parallel trajectory-level success reward and two step-level rewards: Diverse Action Reward and Diverse State Transition Reward, which actively penalize behavioral redundancy and promote broad exploration. Extensive experiments on ALFWorld and ScienceWorld show that DPEPO achieves state-of-the-art (SOTA) success rates, while maintaining comparable efficiency to strong sequential baselines. (Code is available at https://github.com/LePanda026/Code-for-DPEPO)

Junshuo Zhang, Chengrui Huang, Feng Guo, Zihan Li, Ke Shi, Menghua Jiang, Jiguo Yu, Shuo Shang, Shen Gao• 2026

Related benchmarks

TaskDatasetResultRank
Interactive Task CompletionAlfWorld
Pick Success Rate100
45
Interactive Task CompletionScienceWorld
L0 Score66.6
13
Embodied Task CompletionALFWorld (Mixed)
In-Domain Completion Rate98.6
7
Showing 3 of 3 rows

Other info

Follow for update