---
title: "AdaGround: Token-Normalized Adaptive Zooming with Verifier-in-the-Loop Refinement for GUI Grounding"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2610.36874v1"
permalink: "https://feed.astralor.com/items/itm_629dd76ec42f38c23ffe8c54"
published: "2026-10-27T09:12:35.000Z"
score: 56
channel: "tech"
tags: ["多模态","AdaGround","Token-Normalized Adaptive Zooming","Verifier-in-the-Loop Refinement","GUI Grounding GUI","However","GUI"]
---

# AdaGround: Token-Normalized Adaptive Zooming with Verifier-in-the-Loop Refinement for GUI Grounding

## At a glance

GUI grounding on high-resolution screens fails mostly on small targets, and zooming in is a common training-free remedy. However, most existing zooming pipelines do not match the zoom to the target's size, accept the zo…

## Key points

- GUI grounding on high-resolution screens fails mostly on small targets, and zooming in is a common training-free remedy.
- However, most existing zooming pipelines do not match the zoom to the target's size, accept the zoomed answer unconditionally and never revisit a wrong answer.
- To address these issues, we propose AdaGround, a training-free framework with three components: (1) Token-Normalized Adaptive Zooming, which sets the magnification, field of view and zoom depth of each sample from the m…

## Full text

GUI grounding on high-resolution screens fails mostly on small targets, and zooming in is a common training-free remedy. However, most existing zooming pipelines do not match the zoom to the target's size, accept the zoomed answer unconditionally and never revisit a wrong answer. To address these issues, we propose AdaGround, a training-free framework with three components: (1) Token-Normalized Adaptive Zooming, which sets the magnification, field of view and zoom depth of each sample from the model's own estimate of the target size; (2) Verifier-in-the-Loop Refinement, which uses one marked-click Yes/No verifier to keep or reject each zoom, choose between zoom chains, trigger re-grounding and, with an existence query, decide when to abstain; and (3) Compositional Structure Grounding, which composes spreadsheet cells from their headers. On Qwen3-VL models, AdaGround surpasses state-of-the-art training-free zooming methods and many trained GUI models, improving accuracy over the backbone by 7.4 to 20.6 points on OSWorld-G and ScreenSpot-Pro, and its gains carry over to held-out UI-Vision. Code is available at https://github.com/bcwang-sjtu/AdaGround

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_629dd76ec42f38c23ffe8c54