# ARC推理范式提升机器人基础模型零样本性能82.2%

> 原文链接：https://robotics.hellobike.com/papers/2026-10-08-2610-12386
> 频道：papers | 更新日期：2026-10-08 | 标签：auto, arxiv
> 核心摘要：ARC框架通过因果推理轨迹与自动标注管线，无需新增数据即可增强RFM零样本能力。适配π0.5后真机任务成功率提升82.2个百分点，在RoboLab-Reasoning-50基准涨幅达50%，确立无需扩大训练规模的性能新SOTA。

# ARC推理范式提升机器人基础模型零样本性能82.2%

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[ARC: A Reasoning Recipe for Robot Foundation Models](https://arxiv.org/abs/2610.12386)（arXiv 2610.12386）

## 核心内容

针对机器人基础模型（RFM）依赖大规模数据与算力的瓶颈，研究提出ARC推理配方，证明正确的推理策略可显著提升现有SOTA模型的零样本任务性能。该方法包含因果推理轨迹、可扩展自动标注管线及模型适配策略三大要素。实验显示，ARC使π0.5在真机任务成功率提升82.2个百分点，在RoboLab-Reasoning-50基准上最高提升50个百分点，刷新了不增加机器人演示数据条件下的性能纪录。

技术上，ARC要求推理轨迹必须锚定机器人下一步动作并解释其因果结构。研究基于DROID数据集自动构建ARC-Trace-DROID，免除了新数据采集成本。针对π0.5等VLA模型和Cosmos3-Nano-Policy等WAM模型，设计了差异化的微调与推理适配方案。该方案在RoboLab-120和MolmoSpaces基准上均达到新SOTA，验证了通过结构化推理增强而非单纯堆砌参数来提升具身智能控制能力的有效性。

## 关键看点

- **真机性能跃升**：ARC适配π0.5模型后，真实机器人任务成功率提升82.2个百分点，无需额外采集演示数据或进行基础规模训练
- **因果推理轨迹**：推理链锚定下一步动作并解释因果结构，阐明动作合理性及预期效果，为模型提供可解释的控制逻辑支撑
- **自动化数据构建**：从现有DROID数据集自动生成ARC-Trace-DROID，实现推理标注管线可扩展，规避高昂的新数据采集成本
- **基准测试SOTA**：在RoboLab-Reasoning-50上提升达50个百分点，同时在RoboLab-120与MolmoSpaces刷新最优记录

## 工程与产业观察

ARC证明了在不扩充数据集与算力前提下，通过注入结构化因果推理轨迹即可大幅释放存量RFM潜力。这为具身智能研发提供了“推理增强”新范式，提示工程团队应重视高质量推理数据的自动化生成与模型架构适配，而非盲目追求数据规模，对降低真机部署门槛具有重要参考价值。

## 英文摘要 (Abstract)

The prevailing approach to improving robot foundation models (RFMs) relies on larger models, more robot demonstrations, and costly training at scale. We show that there exists an effective and efficient complementary approach: the right reasoning recipe can substantially improve the zero-shot task performance of existing state-of-the-art RFMs. We refer to this recipe as ARC. It consists of three key ingredients: a reasoning trace, a scalable automatic labeling pipeline, and a strategy for adapting pretrained RFMs to use these traces for control. First, we find that effective reasoning traces should be grounded in the robot's next action and explain its causal structure: why the action is appropriate and what effect it should produce. Second, we show that these traces can be generated automatically from existing demonstrations, enabling us to construct ARC-Trace-DROID from DROID without collecting new robot data. Third, we show how state-of-the-art VLAs such as $π_{0.5}$ and WAMs such as Cosmos3-Nano-Policy can learn to use these traces for control, with fine-tuning and inference tailored to each model's architecture and capabilities. Using ARC, we obtain gains in zero-shot RFM performance that, to our knowledge, are unprecedented without additional robot demonstrations or foundation-scale training. The adapted models establish a new state of the art on RoboLab-120 and MolmoSpaces, with gains of up to 50 percentage points on RoboLab-Reasoning-50. On real robots, ARC improves $π_{0.5}$'s task success by 82.2 percentage points. Project website: https://arc-robot-reasoning.github.io/

作者：Gokul Puthumanaillam, Tao Sun, Elie Aljalbout, Moritz Reuss, Zhaoshuo Li, Fabio Ramos, Ankit Goyal, Jenai Xuning Yang
:::

## 原始参考来源 (Sources)

- [arXiv 2610.12386：ARC: A Reasoning Recipe for Robot Foundation Models](https://arxiv.org/abs/2610.12386)