---
title: "外科世界-动作模型：赋能数据高效的手术机器人学习"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.11204v1"
permalink: "https://feed.astralor.com/items/itm_409408d0cf01341cd855486d"
published: "2026-08-11T17:59:13.000Z"
score: 75
channel: "ai"
tags: ["Cosmos Policy","Surgical WAM","世界模型","手术机器人","视频预训练"]
---

# 外科世界-动作模型：赋能数据高效的手术机器人学习

## AI 点评

> 这条论文最值得关注的地方在于用视频预训练换掉一部分昂贵动作标签，思路简单但效果直接，对正在做手术机器人数据采集的团队尤其有启发。不过模拟任务的成功率不等同于真实手术环境的表现，下一步要看它在临床数据上的迁移能力。

## 速览

手术机器人策略学习受限于带动作标注的演示数据稀缺，因为遥操作轨迹如 dVRK 的同步视频-运动学数据采集成本高昂，而内镜视频相对廉价丰富。为此，研究者提出 Surgical WAM，一种基于 Cosmos Policy 的统一生成模型，联合预测未来内镜观察与可执行动作块。该模型先从无动作视频学习视觉动态，再在固定动作标签预算上微调，部署时作为闭环滚动时域控制器。在四个模拟手术任务中，视频预训练将平均成功率从 63.5% 提升至 77.8%，尤其在接触密集和双人任务上效果显著。

## 关键点

- 提出 Surgical WAM，一种基于 Cosmos Policy 的世界-动作模型，联合预测内镜视频与手术机器人动作块。
- 利用无动作内镜视频预训练视觉动态先验，显著降低对昂贵动作标签的依赖。
- 在四个模拟手术任务上，平均成功率从 63.5% 提升至 77.8%，PegTransfer 绝对提高 20 个百分点。
- 验证了视频预训练作为扩展手术机器人学习的数据高效路径。

## 全文

Learning reliable surgical manipulation policies is bottlenecked by the scarcity of action-labeled demonstrations: teleoperated surgical robot (e.g., dVRK) trajectories with synchronized kinematics are costly to collect, while surgical tasks demand precise contact handling, long-horizon reasoning, and bimanual coordination. Endoscopic video is comparatively inexpensive and abundant relative to synchronized video--kinematics trajectories, and a natural way to exploit it is to learn world models of surgical scenes. However, existing surgical world models use video primarily for simulation or policy evaluation, and rarely translate the learned dynamics into closed-loop control. This gap raises our central question: under a fixed budget of action-labeled demonstrations, does action-free video pretraining improve closed-loop surgical manipulation? To answer it, we introduce the Surgical World-Action Model (Surgical WAM), a unified generative model built on Cosmos Policy that jointly predicts future endoscopic observations and executable surgical robot action chunks. Surgical WAM first learns surgical visual dynamics from action-free video and is then fine-tuned on the fixed action-labeled budget; at deployment, it acts as a closed-loop, receding-horizon controller that executes a short prefix of each predicted action chunk and replans from the resulting observation. On a suite of four simulated surgical manipulation tasks, video pretraining improves the average success rate from 63.5% to 77.8%, including an absolute gain of 20 percentage points on PegTransfer, with the largest improvements on contact-rich and bimanual tasks. These results demonstrate that action-free video provides transferable visual dynamics priors for learning surgical robot control with limited action supervision, positioning data-efficient video pretraining as a practical path toward scaling up surgical robot learning.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_409408d0cf01341cd855486d