---
id: 20260514-T0-04
title: "ReVision：削减90%视觉冗余token，让计算机操作代理跑得更快更省"
title_en: "ReVision: Cutting 90% Visual Redundancy Tokens for Faster Computer-Use Agents"
url: https://ai.daily.yangsir.net/daily/20260514-T0-04
issue_date: 2026-05-14
publish_date: 2026-05-13T04:00:00.000Z
category: research
source_name: "arXiv cs.CL (NLP)"
source_url: https://arxiv.org/abs/2605.11212
---

# ReVision：削减90%视觉冗余token，让计算机操作代理跑得更快更省

arXiv 论文（2605.11212）提出 ReVision 方法，解决计算机操作代理（CUA）中视觉 token 开销过大的问题。传统 CUA 代理每次截图都编码大量视觉 token，交互轨迹越长成本越高。ReVision 通过减少时序视觉冗余来优化这一过程，在保留关键视觉信息的同时大幅降低 token 消耗。研究结果表明，该方法能显著降低计算成本，同时保持代理操作精度。开发者可以用更低的 API 成本运行长时间 GUI 自动化任务。

## English Version

**ReVision: Cutting 90% Visual Redundancy Tokens for Faster Computer-Use Agents**

A new arXiv paper (2605.11212) proposes ReVision to address excessive visual token costs in computer-use agents (CUAs). Traditional CUAs encode large numbers of visual tokens per screenshot, making longer interactions prohibitively expensive. ReVision reduces temporal visual redundancy while preserving critical visual information. Results show substantial cost reduction without sacrificing operational accuracy. Developers can run long-duration GUI automation tasks at significantly lower API costs.

---

**来源**：[arXiv cs.CL (NLP)](https://arxiv.org/abs/2605.11212)

**详情页**：https://ai.daily.yangsir.net/daily/20260514-T0-04

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*