# TaskAnchor lets VLA use execution history to address task-state confusion in long-horizon manipulation tasks

- Source: Xbotics具身智能实验室 · 微信公众号
- Published: 2026-10-22T09:00:41.000Z
- RoboHot: https://robohot.news/en/items/ewa9iimhfwkp5u2ufh7vkb1pt
- Original: https://mp.weixin.qq.com/s?__biz=MzkyMTc1NTI3Ng==&mid=2247509447&idx=1&sn=6d8fd18175fe3be629943a6a7324908a&chksm=c02aee470c881ea48b0af1491f0e7edf702b0cfee226c6da109dab6bba202562b17686763b3c&scene=126&sessionid=1792665994#rd

## Summary

TaskAnchor connects execution history to VLA's existing language and vision interfaces through a lightweight adapter, aiming to alleviate task-state confusion in long-horizon manipulation. It uses a dual-path design: a normalized task-state coordinate that describes semantic execution stages, and a visual residual based on context-adaptive test-time training (CA-TTT) that preserves historical details.