arXiv：cs.AI· Kaiming Liu, Xuanyu Lei, Ziyue Wang, Peng Li, Yang Liu·· 23 小时前
ALIGN：通过自动接口生成实现 LLM 智能体与环境对齐
Agent-Environment Alignment via Automated Interface Generation
AI 导读
针对 LLM 智能体内部预期与环境实际状态转换不匹配导致的"智能体-环境不对齐"问题，研究者提出 ALIGN 框架，通过自动生成接口丰富环境静态信息与逐步观测。该接口以轻量级 wrapper 形式实现，无需改动智能体逻辑或环境代码，在 ALFWorld 上成功率最高提升 45.67%，并在具身任务、网页导航和工具使用等领域取得一致提升，可跨智能体架构与 LLM 骨干泛化而无需重新生成接口。
正文
Abstract:Large language model (LLM) agents have shown impressive reasoning capabilities in interactive decision-making tasks. These agents interact with environment through intermediate interfaces, such as predefined action spaces and interaction rules, which mediate the perception and action. However, mismatches often happen between the internal expectations of the agent regarding the influence of its issued actions and the actual state transitions in the environment, a phenomenon referred to as \textbf{agent-environment misalignment}. While prior work has invested substantially in improving agent strategies and environment design, the critical role of the interface still remains underexplored. In this work, we empirically demonstrate that agent-environment misalignment poses a significant bottleneck to agent performance. To mitigate this issue, we propose \textbf{ALIGN}, an \underline{A}uto-A\underline{l}igned \underline{I}nterface \underline{G}e\underline{n}eration framework that alleviates the misalignment by enriching the interface. Specifically, the ALIGN-generated interface enhances both the static information of the environment and the step-wise observations returned to the agent. Implemented as a lightweight wrapper, this interface achieves the alignment without modifying either the agent logic or the environment code. Experiments across multiple domains including embodied tasks, web navigation and tool-use, show consistent performance improvements, with up to a 45.67\% success rate improvement observed in ALFWorld. Meanwhile, ALIGN-generated interface can generalize across different agent architectures and LLM backbones without interface regeneration. Code and experimental results are available at this https URL.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2505.21055 [cs.AI] |
| (or arXiv:2505.21055v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2505.21055 arXiv-issued DOI via DataCite |
Submission history
From: Kaiming Liu [view email]
[v1]
Tue, 27 May 2025 11:44:50 UTC (2,877 KB)
[v2]
Tue, 13 Oct 2026 06:45:18 UTC (1,692 KB)
来源：arXiv：cs.AI · arxiv.org