跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.02053 · v1

Learning to Look: A Foundation for Visual Understanding

Wenshuai ZhaoYihao WangBenjamin RobsonJuho KannalaArno Solin
cs.CV
首次提交 2026-11-02 00:57
版本更新 2026-11-02 00:57
雷达观测 2026-11-04 23:43
arXiv ↗PDF ↗

摘要

Vision-language models (VLMs) suffer from visual hallucination, such as saying an object that is not present in an image. We ask a more fundamental question: whether a VLM is even aware of where it is looking before understanding the semantic meaning in the image. We formulate this question as a LOOK task: identifying which parts of its visual input were available to the model. We construct the task in a self-supervised way without manually labeling the available region to the model. Our experiments show that existing VLMs fail to localize the visible region. We further show that fine-tuning VLMs on the proposed LOOK task improves a model's awareness of visual input.

作者备注：15 pages, 7 figures

查看 arXiv 原文 ↗阅读 PDF ↗

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存，登录后共享已有结果。

版本差异

目前只观测到一个版本，无法比较变化。之后收录新版本时，会在这里对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本；首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-02 00:57 · 最新观测版本

Learning to Look: A Foundation for Visual Understanding

Vision-language models (VLMs) suffer from visual hallucination, such as saying an object that is not present in an image. We ask a more fundamental question: whether a VLM is even aware of where it is looking before understanding the semantic meaning in the image. We formulate this question as a LOOK task: identifying which parts of its visual input were available to the model. We construct the task in a self-supervised way without manually labeling the available region to the model. Our experiments show that existing VLMs fail to localize the visible region. We further show that fine-tuning VLMs on the proposed LOOK task improves a model's awareness of visual input.

查看此版本解读 →查看此版本原文 ↗