AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

DictAS: A Framework for Class-Generalizable Few-Shot Anomaly Segmentation via Dictionary Lookup

Zhen Qu, Xian Tao, Xinyi Gong, ShiChen Qu, Xiaopei Zhang, Xingang Wang, Fei Shen, Zhengtao Zhang, Mukesh Prasad, Guiguang Ding

ICCV 2025 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Query Generator (ablation)DictAS · MVTecAD 4-shotpixel AP: 63.5 -> 66.8Improved
add Key Generator (ablation)DictAS · MVTecAD 4-shotpixel AP: 63.8 -> 66.8Improved
add Value Generator (ablation)DictAS · MVTecAD 4-shotpixel AP: 64.2 -> 66.8Improved
add Contrastive Query Constraint (ablation)DictAS · MVTecAD 4-shotpixel AP: 64.6 -> 66.8Improved
add Text Alignment Constraint (ablation)DictAS · MVTecAD 4-shotpixel AP: 65.0 -> 66.8Improved
add both query discrimination regularizers (ablation)DictAS · MVTecAD 4-shotpixel AP: 63.7 -> 66.8Improved
use sparse lookup instead of maximum lookup (ablation)DictAS · MVTecAD 16-shotpixel AP: 60.6 -> 68.5Improved
use sparse lookup instead of dense lookup (ablation)DictAS · MVTecAD 16-shotpixel AP: 63.8 -> 68.5Improved
use sparse lookup instead of dense lookup (ablation)DictAS · MVTecAD 1-shotpixel AP: 60.2 -> 61.1Improved
raise input resolution from 336x336 to 518x518 (ablation)DictAS · MVTecAD 4-shotpixel AP: 66.8 -> 68.7Improved
use ViT-L-14-336 instead of ViT-B-16 backbone (ablation)DictAS · MVTecAD 4-shotpixel AP: 64.8 -> 66.8Improved
train on natural-scene Ade20K instead of VisA (ablation)DictAS · MVTecAD 4-shotpixel AP: 66.8 -> 66.5No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy