AI Paper Atlas

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

Concept Regions Matter: Benchmarking CLIP with a New Cluster-Importance Approach

Aishwarya Agarwal, Srikrishna Karanam, Vineet Gandhi

CVPR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Open-vocabulary alignment from VLMs · 1 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
use cluster-based concept masking (CCI) instead of Grad-ECLIP for attributionCLIP ViT-B/16 · ImageNet-1K valDeletion AUC Top-1 (lower better): 0.2464 -> 0.1809Improved
use CCI instead of Grad-ECLIP on COCO image retrievalCLIP ViT-B/16 · MS COCO IRDeletion AUC @5 (lower better): 0.2670 -> 0.1056Improved
use CCI instead of Grad-ECLIP for insertionCLIP ViT-B/16 · ImageNet-1K valInsertion AUC Top-5: 0.5993 -> 0.6518Improved
add Grad-ECLIP lambda_i weighting (ablation)Grad-ECLIP · ImageNet-1K valDeletion AUC Top-1 (lower better): 0.2535 -> 0.2464Improved
vary background with synthetic editing (COVAR Bg-varied)CLIP ViT-B/16 · COVARBG-Er (%): 6.7 -> 15.6Hurt
reduce object scaleCLIP variants · COVARaccuracy: 62.2 -> 32.6 (ViT-L/14 DataComp-1B)Hurt
reduce object scaleViT-H-qgelu 378px · COVARaccuracy: 65.2 -> 39.3Hurt
crop, horizontal flip or translation of imagesCLIP variants · COVARaccuracy: 55.8 -> 55.2 (H-flip), 55.8 (Translate), 59.4 (Crop), ViT-B/32 DataComp-1BNo clear effect
use smaller patch size (B/16 vs B/32) with DataComp-1B (ablation)CLIP ViT-B · COVAR ScaleBG-Er (%): 50.7 -> 30.5Improved
raise input resolution to 512px (ablation)ViT-B-SigLIP2 · COVARBG-Er under scale: 39.3% BG-Er under scale; Bg-varied accuracy 64.9No clear effect
scale up model sizeCLIP · COVAR ScaleBG-Er (%): 32.7 (ViT-bigG)No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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