AI Paper Atlas

Research map / Label-efficient and robust learning / Noisy labels

CAPro: Webly Supervised Learning with Cross-modality Aligned Prototypes

Yulei Qin, Xingyu Chen, Yunhang Shen, Chaoyou Fu, Yun Gu, Ke Li, Xing Sun, Rongrong Ji

NeurIPS 2023 · Label-efficient and robust learning › Noisy labels · approach: Crowdsourced label aggregation · 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
add CAPro cross-modality aligned prototypes and collective bootstrapping over vanillaR50 vanilla · WebVision1ktop-1 accuracy: 72.6 -> 74.2Improved
add CAPro over vanillaR50 vanilla · ImageNet1ktop-1 accuracy: 67.0 -> 68.0Improved
add CAPro over vanillaR50 vanilla · NUS-WIDE (Web)mAP: 38.3 -> 48.0Improved
add CAPro over vanillaR50 vanilla · Open-set recognition, WebVisionC-F1: 54.6 -> 62.2Improved
add text encoding plus the paper's text enhancement (MiniLM) (ablation)CAPro · Google500top-1 accuracy: 71.5 -> 75.5Improved
replace VSGraph text enhancement with the paper's enhancement (MiniLM) (ablation)CAPro · ImageNet500top-1 accuracy: 66.9 -> 71.5Improved
replace MiniLM text encoder with XLNet (ablation)CAPro with text enhancement · Google500top-1 accuracy: 75.5 -> 75.4No clear effect
replace MiniLM text encoder with GPT-Neo (1.3B) (ablation)CAPro with VSGraph enhancement · Google500top-1 accuracy: 72.0 -> 72.0No clear effect
add collective bootstrapping as reference provider (ablation)CAPro with text enhancement · Google500top-1 accuracy: 75.5 -> 76.0Improved
use SCC as reference provider instead of CB (ablation)CAPro · Google500top-1 accuracy: 75.5 -> 73.8Hurt
use NCR as reference provider instead of CB (ablation)CAPro · NUS-WIDEO-F1: 45.4 -> 43.4Hurt
add mix-up on top of collective bootstrapping (ablation)CAPro · Google500top-1 / top-5 accuracy: 76.0 -> 76.5 top1; 91.3 -> 91.1 top5Mixed

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