# CausGround框架：非参数因果接地提升具身规划成功率至53%

> 原文链接：https://robotics.hellobike.com/papers/2026-10-21-2610-29702-0a689330
> 频道：papers | 更新日期：2026-10-21 | 标签：auto, arxiv, hf-daily-papers
> 核心摘要：针对多模态大模型物理交互规划难题，CausGround提出非参数因果接地框架，将经验蒸馏为可复用动作因果知识。在VirtualHome与RoboCasa基准测试中，任务成功率分别从24.9%升至53.3%、43.0%升至57.0%，无需参数更新即可支持长程操作。

# CausGround框架：非参数因果接地提升具身规划成功率至53%

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[CausGround: Causal Interaction Grounding for Adaptive Embodied Planning](https://arxiv.org/abs/2610.29702)（HF Daily Papers · arXiv 2610.29702）

## 核心内容

针对多模态大模型（MLLMs）在具身智能中缺乏环境依赖型前置条件与效应认知的问题，研究团队提出CausGround非参数因果交互接地框架。该方法无需昂贵的环境特定训练，通过将交互经验蒸馏为以动作为中心的因果知识，解决非结构化经验难以复用的痛点。在VirtualHome和RoboCasa仿真基准及SWITCH推理测试中，该框架显著提升了多种MLLM骨干网络的规划性能，验证了其在无参数更新下跨环境长程规划的有效性。

技术上，CausGround识别动作前置条件与效应，构建包含物体属性及关系的因子化世界状态表征。规划阶段将任务相关因果依赖锚定于当前观测，支持基于前置条件的逆向推理与基于效应的正向推理，以优化动作选择与多步规划。实测数据显示，VirtualHome视觉高层规划平均成功率从24.9%提升至53.3%，RoboCasa闭环规划成功率从43.0%增至57.0%，SWITCH交互推理准确率由25.8%升至31.9%，并在真实机器人序列操作中成功实现从目标到可执行动作序列的逆向推导。

## 关键看点

- **非参数因果框架**：将交互经验蒸馏为动作中心因果知识，无需环境特定训练或模型参数更新即可适配新场景
- **规划成功率跃升**：VirtualHome视觉高层规划平均成功率从24.9%提升至53.3%，RoboCasa闭环规划达57.0%
- **双向推理机制**：基于因子化世界状态表征，结合观测进行前置条件逆向推理与效应正向推理，优化多步决策
- **交互推理增强**：在SWITCH基准上平均准确率从25.8%升至31.9%，有效改善动作选择、错误恢复与验证能力

## 工程与产业观察

CausGround证明了在不微调大模型前提下，通过结构化因果知识检索与推理弥补MLLM物理常识缺失的可行性。其非参数化设计降低了具身智能适配新环境的边际成本，为解决Sim-to-Real鸿沟及长程任务规划提供了轻量级工程范式，对构建通用操作Agent具有重要参考价值。

## 英文摘要 (Abstract)

Multimodal large language models (MLLMs) show promising reasoning capabilities, but reliable physical interaction requires planning that accounts for when actions are executable and how they change the world. These preconditions and effects are environment-dependent and often not captured by MLLMs, making adaptation challenging. Yet existing approaches either require costly environment-specific training or rely on unstructured interaction experience that is difficult to reuse. Therefore, we introduce CausGround, a non-parametric causal interaction grounding framework that distills interaction experience into reusable action-centered causal knowledge. CausGround identifies action preconditions and effects and organizes them in a factorized world-state representation capturing object attributes and inter-object relations. At planning time, task-relevant causal dependencies are grounded in current observations to support backward reasoning over preconditions and forward reasoning over effects for action selection and multi-step planning. We evaluate CausGround on simulated high-level planning, interaction reasoning, and real-world robot manipulation. On adapted VirtualHome and RoboCasa benchmarks, CausGround improves vision-based high-level planning across MLLM backbones, increasing average task success from 24.9% to 53.3% on VirtualHome and closed-loop planning success from 43.0% to 57.0% on RoboCasa. On SWITCH, average interaction reasoning accuracy increases from 25.8% to 31.9%, indicating improved action selection, recovery, and verification. Moreover, real-world experiments show that CausGround reasons backward from MLLM-interpreted goals to derive executable high-level action sequences for sequential manipulation. These results demonstrate that reusable action preconditions and effects support long-horizon embodied planning across diverse environments without parameter updates.

Hugging Face Daily Papers 热度：1 upvotes
作者：Yijun Shen, Diqi He, Jinbang Guo, Wanyue Zhang, Jiaming Mi, Huan Yang, Zhongyuan Wang, Zhiwei Yu, Kai Zhang
:::

## 原始参考来源 (Sources)

- [HF Daily Papers · arXiv 2610.29702：CausGround: Causal Interaction Grounding for Adaptive Embodied Planning](https://arxiv.org/abs/2610.29702)