AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Caption Bottleneck Models

Seref Baris Cagliyan, Umut Ozdemir, Merve Tapli, Emre Akbas

ECCV 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Concept bottleneck models · 0 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
inject top-ranked class concept into caption (human intervention) (ablation)CaBM · CUB-200 intervention subsetaccuracy (%): 41.7 -> 55.5 (+13.8)Improved
inject top-ranked class concept into caption (ablation)CaBM · Flowers-102 intervention subsetaccuracy (%): 60.6 -> 72.8Improved
inject top-ranked class concept into caption (ablation)CaBM · CIFAR-100 intervention subsetaccuracy (%): 62.3 -> 75.2Improved
replace LLM-generated concepts with CaBM-discovered concepts in VLG-CBM (ablation)VLG-CBM · CUB-200ANEC-5 accuracy: 75.79 -> 76.25Improved
replace LLM-generated concepts with CaBM-discovered concepts in VLG-CBM (ablation)VLG-CBM · CIFAR-100ANEC-5 accuracy: 65.73 -> 66.13Improved
use text-only caption bottleneck instead of image-domain CBM (ViT)CaBM vs HybridCBM · CUB-200top-1 accuracy: 84.25 -> 76.92Hurt
use text-only caption bottleneck instead of CNN-backbone CBMsCaBM vs PS-CBM · ImageNet-1Ktop-1 accuracy: 74.00 -> 75.12Improved
use text-only caption bottleneckCaBM · Flowers-102top-1 accuracy: 97.90 (PS-CBM) -> 86.57Hurt
use discovered concepts rather than LLM-proposed set (concept quality)CaBM vs VLG-CBM · CUB-200Semantics: 0.86 -> 0.92Improved

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