跳到正文
arXiv cs.AI· Alin-Ionut Popa, Tuo Zhang, Dimitrios Dimitriadis·· 10 小时前AI 评分38

SEE:无需训练,让冻结的 VLM 重新审视自己指向的区域

SEE: Unlocking the Hidden Perception of Frozen VLMs

AI 导读

研究者提出免训练方法 SEE(Self-Elicited Evidence),在冻结 VLM 初次回答置信度低时,让模型自行提出细节可能出现的区域,在放大裁剪图上重新作答,并与原始预测做置信度加权投票。

正文 · 原文

View PDF HTML (experimental)

Abstract:Vision-language models (VLMs) grasp the overall content of an image but often miss the fine details a question hinges on, such as a small object in a large photo or a number in a dense infographic. Yet even when their answer is wrong, the region they point to often contains the evidence. We introduce Self-Elicited Evidence (SEE), a training-free method that lets a frozen VLM look again where it points. When the initial answer has low confidence, SEE prompts the model to propose regions where the details may appear, answers the question on these enlarged crops, and aggregates the answers with the original prediction through confidence-weighted voting. The model also answers with its own transcript, which helps on text-rich images. SEE needs no detector or OCR engine. Across five backbones, SEE improves every small-object benchmark by 2.3-13.1 points and outperforms five training-free baselines on small objects. With SEE, Qwen3.5-4B surpasses the direct answer of Qwen3.5-9B and matches the 6 times larger Qwen3.5-27B.
Subjects: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.23962 [cs.CV]
  (or arXiv:2611.23962v1 [cs.CV] for this version)
  https://doi.org/10.48550/arXiv.2611.23962

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Alin Popa [view email]
[v1] Wed, 18 Nov 2026 06:53:22 UTC (14,292 KB)

来源:arXiv cs.AI · arxiv.org