arXiv cs.AI· Ye Li, Haoran Ding, Huanan Liu, Yuan Meng, Yutian Shen, Yuansong Wang, Kangye Ji, Hanyun Cui, Hongbo Duan, Chen Tang, Jiawei Li, Shu-Tao Xia, Zhi Wang·· 3 小时前AI 评分38
GUIFlow：面向 GUI 智能体的时序冗余感知增量推理框架
GUIFlow: Temporal Redundancy-Aware Incremental Inference for GUI Agents
AI 导读
清华大学等团队在 arXiv 提出 GUIFlow，一个面向多图像 GUI 智能体交互的全流程增量推理框架，通过历史 KV 状态位置对齐注入、视觉编码器与语言模型的选择性更新，以及以早前动作响应为推测草稿并由目标模型验证，实现跨步骤复用。
正文 · 原文
Authors:Ye Li, Haoran Ding, Huanan Liu, Yuan Meng, Yutian Shen, Yuansong Wang, Kangye Ji, Hanyun Cui, Hongbo Duan, Chen Tang, Jiawei Li, Shu-Tao Xia, Zhi Wang
Abstract:GUI agents built on vision-language models interact with continuously evolving interfaces, yet conventional inference repeatedly recomputes representations of largely overlapping observations and histories. This mismatch makes multi-step interaction costly, while changes in context, interface content, and action arguments prevent straightforward cache reuse. We introduce GUIFlow, a framework for full-pipeline incremental inference in multi-image GUI interaction. Our central insight is that advancing a decision need not require rebuilding all supporting representations: each interaction step can instead selectively update computational states maintained across the trajectory. To support new decisions as the context evolves, GUIFlow adapts how past computation is reused: historical KV states are position-aligned and injected into the current context; change signals guide selective updates of current-observation states in the vision encoder and language model; and earlier action responses serve as speculative drafts verified by the target model. Together, these mechanisms enable cross-step reuse throughout visual encoding, prefill, and decoding without additional training, a separate draft model, or changes to the model-visible inputs and action interface. Experiments demonstrate up to 3.39x geometric-mean speedup across offline multi-image benchmarks and up to 3.49x in online, closed-loop interaction on OSWorld, while maintaining task performance broadly comparable to the dense baseline.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.34959 [cs.CV] |
| (or arXiv:2610.34959v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.34959 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ye Li [view email]
[v1]
Mon, 26 Oct 2026 07:25:39 UTC (37,474 KB)
来源：arXiv cs.AI · arxiv.org