EnvACE: Internalizing Environment Dynamics via World Rehearsal for Agentic Reinforcement Learning

Source

• Source name: arXiv
• Source kind: paper
• Authors / organization: Zishan Xu, Zhiyuan Yao, Yuxin Chen, Yifu Guo, Zhengxi Lu, Yuquan Lu, Jinyang Huang, Yan Xu, Yasheng Wang, Weinan Zhang, Xingshan Zeng, Weiwen Liu
• Venue: arXiv: cs.AI
• Published: 2026-08-06
• Collected: 2026-08-07
• URL: http://arxiv.org/abs/2608.06197v1
• PDF: https://arxiv.org/pdf/2608.06197v1
• DOI: UNKNOWN

Plain-Language Brief

The researchers introduce EnvACE, a method for reinforcement learning that replaces external environment interactions during training with world rehearsal. This method allows the policy to internalize the relationship between actions and their environment responses, leading to strong and transferable performance across various benchmarks. At test time, the internalized world model enables private rehearsal before committed execution, yielding further gains.

What the Document Says

• The paper introduces EnvACE, an agentic reinforcement learning method that uses world rehearsal during training instead of external environment interactions or costly simulators.
• The policy in EnvACE alternates between acting and rehearsal: it generates a tool call, plays the role of the environment to produce the response induced by that action, and conditions subsequent decisions on the rehearsed response.
• Both roles are jointly optimized end-to-end using task-success rewards.
• Through world rehearsal, the policy internalizes the relationship between actions and their environment responses in its parameters, yielding an agent world model that supports decision making.
• Across several benchmarks including BFCL-v4, tau^2-Bench, VitaBench, and FinMCP-Bench, EnvACE achieves strong and transferable performance, outperforming environment-scaling baselines in the overall evaluation.
• Controlled studies show that world rehearsal improves policy learning across model scales.
• At test time, the internalized world model enables private rehearsal before committed execution, yielding further gains under a moderate rehearsal budget without additional external interaction.

Key Concepts Mentioned

• reinforcement learning
• world rehearsal
• agent world model
• policy learning
• benchmarks

Claims Made by the Document

• EnvACE achieves strong and transferable performance across various benchmarks.
• World rehearsal improves policy learning across model scales.

Evidence Used

• Across BFCL-v4, tau^2-Bench, VitaBench, and FinMCP-Bench, EnvACE outperforms environment-scaling baselines.
• Controlled studies show consistent improvements in policy learning with world rehearsal.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Training large language model agents for long-horizon tool use typically relies on interactions with real or synthesized executable environments, whose construction and verification are costly, or on external simulators that are difficult to ground. We introduce EnvACE, an agentic reinforcement learning method that replaces external environment interaction during training with world rehearsal. The policy alternates between acting and rehearsal: it first generates a tool call, then plays the role of the environment to produce the response induced by that action, and conditions subsequent decisions on the rehearsed response. Both roles are jointly optimized end-to-end using task-success rewards. Through world rehearsal, the policy internalizes the relationship between actions and their environment responses in its parameters, yielding an agent world model that directly supports decision making. Across BFCL-v4, tau^2-Bench, VitaBench, and FinMCP-Bench, EnvACE achieves strong and transferable performance, outperforming environment-scaling baselines in the overall evaluation. Controlled studies further show that world rehearsal consistently improves policy learning across model scales. At test time, the internalized world model enables private rehearsal before committed execution, yielding further gains under a moderate rehearsal budget without additional external interaction. Our findings establish world rehearsal as a new path toward scaling LLM agent training beyond the constraints of external environments. Our code is publicly available at https://github.com/Within-yao/EnvACE.