AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Webly Supervised Concept Expansion for General Purpose Vision Models

Amita Kamath, Christopher Clark, Tanmay Gupta, Eric Kolve, Derek Hoiem, Aniruddha Kembhavi

ECCV 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Human-like visual cognition in VLMs · 42 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 WEB10K web-search data to multi-task trainingGPV-2 · DCE (captioning, out-of-domain)CIDEr-D: 65.4 -> 72.5Improved
add WEB10K web-search data to multi-task trainingGPV-2 · DCE (classification)accuracy: 36.3 -> 45.4Improved
add WEB10K web-search data to multi-task trainingGPV-2 · DCE (localization)mAP: 74.8 -> 75.9Improved
add WEB10K web-search data to multi-task trainingVL-T5 · DCE (captioning, out-of-domain)CIDEr-D: 31.6 -> 45.0Improved
add WEB10K web-search data to multi-task trainingGPV-2 · COCO (VQA test-dev)VQA accuracy: 71.1 -> 71.4No clear effect
add WEB10K web data (closed world, COCO-SCE DETR features)GPV-2 · DCE (captioning, out-of-domain)CIDEr-D: 21.1 -> 33.4Improved
add classifier re-calibration (ablation)GPV-2 · DCE (classification)accuracy: 34.0 -> 45.8Improved
add Conceptual Captions pre-training (ablation)GPV-2 · DCE (captioning, out-of-domain)CIDEr-D: -13.9Improved
use VinVL features instead of DETR features (ablation)GPV-2 · DCE (captioning, out-of-domain)CIDEr-D: -41.7Improved
use WEB10K in multi-task training rather than as a second pre-training round (ablation)GPV-2 · DCE (captioning, out-of-domain)CIDEr-D: -8.2Improved
add language-based localization (LBL) (ablation)GPV-2 · DCE (captioning, out-of-domain)CIDEr-D: -2.4Improved
add WEB10K web data to zero-shot attribute recognitionGPV-2 · VAWmAP: 53.2 -> 52.4No clear effect

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