Towards Reliable and Robust LLM Planning: Symbolic Feedback-Driven Iterative Self-Refinement Framework

面向可靠且稳健的 LLM 规划：基于符号反馈的迭代自优化框架

Abstract: Large language models (LLMs) have attracted widespread attention from academia and industry, yet their deployment raises critical security concerns regarding robustness and reliability. Planning, a core component of intelligent behavior, remains challenging for LLMs, which often produce infeasible or incorrect solutions in long-horizon decision-making tasks due to inherent complexity.

摘要： 大语言模型（LLM）已引起学术界和工业界的广泛关注，但其部署在稳健性和可靠性方面引发了严峻的安全担忧。规划作为智能行为的核心组成部分，对 LLM 而言仍然极具挑战性；由于任务的内在复杂性，LLM 在长程决策任务中往往会产生不可行或错误的解决方案。

In this paper, we propose a symbolic feedback-driven iterative self-refinement framework to enhance the robustness and reliability of LLMs in long-horizon planning. Specifically, a natural language prompting mechanism is introduced to map logical symbols into natural language descriptions, enabling LLMs to better capture task constraints and semantics.

在本文中，我们提出了一种基于符号反馈的迭代自优化框架，旨在增强 LLM 在长程规划中的稳健性和可靠性。具体而言，我们引入了一种自然语言提示机制，将逻辑符号映射为自然语言描述，使 LLM 能够更好地捕捉任务约束和语义。

We further design a symbolic verifier that identifies errors and converts them into corrective instructions interpretable by the LLM, thereby guiding self-refinement. In addition, we leverage a plan recognizer to infer goal reachability, facilitating more effective guidance toward desired goals.

我们进一步设计了一个符号验证器，用于识别错误并将其转换为 LLM 可理解的修正指令，从而指导其进行自优化。此外，我们利用规划识别器来推断目标的可达性，从而更有效地引导模型实现预期目标。

Empirical results demonstrate that the proposed framework consistently improves both feasibility and correctness in long-horizon planning tasks. This highlights its effectiveness in enhancing the reliability of LLM-based planning and potential to enable more trustworthy AI systems.

实证结果表明，该框架在长程规划任务中持续提升了方案的可行性和正确性。这凸显了该方法在增强基于 LLM 的规划可靠性方面的有效性，并展现了其构建更可信 AI 系统的潜力。