---
title: "多模态大语言模型的规则合规视觉空间规划：RuleMaze基准与解耦规划方法"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.20237v1"
permalink: "https://feed.astralor.com/items/itm_dc47b856af3c43484c396efc"
published: "2026-08-20T16:28:28.000Z"
score: 79
channel: "ai"
tags: ["RuleMaze","可解释推理","多模态大语言模型","空间规划","规则遵循"]
---

# 多模态大语言模型的规则合规视觉空间规划：RuleMaze基准与解耦规划方法

## AI 点评

> 这篇论文值得关注：它把MLLM的评测从视觉问答推进到规则约束下的空间规划，直击当前大模型在可控性上的短板。对做具身智能、Agent规划或模型可解释性的研究者，RuleMaze和DMP都是值得深入对比的基线。

## 速览

多模态大语言模型虽兼具语言推理与视觉感知，但在显式或未见过的规则约束下进行视觉空间规划的能力尚待探索。为此，研究者提出RuleMaze基准，要求模型在遵循自然语言规则的同时穿越迷宫，并设计Language-Logic-Function混合方法自动生成规则与验证器。同时提出Disentangled Multimodal Planning (DMP)，将感知、执行与规则验证解耦，提升规则遵循与泛化能力。实验表明DMP显著优于端到端文本规划基线，为可解释的规则化空间规划提供了新思路。

## 关键点

- 提出RuleMaze基准，用于评估MLLM在自然语言规则约束下的视觉空间规划能力。
- 设计Language-Logic-Function Hybridization，自动生成规则并转换为可执行的逻辑验证器。
- 提出Disentangled Multimodal Planning，将感知、执行与规则验证解耦，提升泛化性与可解释性。
- 实验证明DMP在规则遵循和规划成功率上显著优于端到端文本规划基线。

## 全文

Multimodal large language models (MLLMs) combine linguistic reasoning with visual perception, yet their ability to perform visual spatial planning under explicit or previously unseen rule constraints remains underexplored. This setting requires models to jointly understand spatial layouts, interpret natural-language rules, and plan valid actions accordingly. To address this gap, we introduce RuleMaze, a controllable benchmark in which MLLMs must navigate mazes while obeying natural-language rules of varying complexity. RuleMaze isolates rule-compliant spatial planning by requiring accurate perception, rule interpretation, and constrained action planning. To enable scalable and systematic rule construction, we propose Language-Logic-Function Hybridization, which automatically generates natural-language rules and translates them into logical representations and executable validators, eliminating manual rule engineering. To improve rule following and generalization, we introduce Disentangled Multimodal Planning (DMP), which separates perception, execution, and rule verification through interpretable reasoning primitives. By disentangling these components, DMP facilitates systematic generalization to more complex and previously unseen rules, while providing transparent intermediate planning traces. Experiments demonstrate that DMP substantially improves rule compliance and planning success compared to end-to-end textual planning baselines. Overall, RuleMaze establishes a principled benchmark for studying grounded and interpretable rule-based spatial planning in MLLMs. Code is available at https://github.com/oceanflowlab/RuleMaze.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_dc47b856af3c43484c396efc