AI Paper Atlas

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

Attention-Challenging Multiple Instance Learning for Whole Slide Image Classification

Yunlong Zhang, Honglin Li, YUXUAN SUN, Chenglu Zhu, Sunyi Zheng, Lin Yang

ECCV 2024 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 111 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 diversity loss L_d to multiple branch attention (ablation)ACMIL · CAMELYON16 (ViT-S/16 SSL)AUC: 0.943 -> 0.974Improved
add diversity loss L_d to multiple branch attention (ablation)ACMIL · BRACS (ResNet-18)AUC: 0.760 -> 0.754No clear effect
keep STKIM active at test time (ablation)ACMIL · CAMELYON16/BRACS/LBC (ViT-S/16 SSL)F1/AUC: 0.954 -> 0.927 F1 on CAMELYON16Hurt
mask all Top-K instances (p=1.0) in STKIM (ablation)ACMIL · CAMELYON16/BRACS/LBCAUCHurt
use WENO-style masking (95 instances, p=1.0) instead of STKIM default (ablation)ACMIL · CAMELYON16 (ViT-S/16 SSL)AUC: 0.974 -> 0.843Hurt
vary number of masked instances K in STKIM (ablation)ACMIL · CAMELYON16/BRACS/LBCAUC: less than 1.0% AUC differenceNo clear effect
add Multiple Branch Attention to ABMIL (ablation)ABMIL · CAMELYON16 (ResNet-18)AUC: 0.790 -> 0.850Improved
add STKIM to ABMIL (ablation)ABMIL · BRACS (ResNet-18)AUC: 0.723 -> 0.789Improved
combine MBA with STKIM (ablation)ACMIL · CAMELYON16/BRACS/LBCAUC: 39 of 45 configurations exceed MBA/STKIM-alone lineImproved
apply ACMIL on top of multi-head attention instead of gated attentionMHA baseline · BRACS (ResNet-18)AUC: 0.738 -> 0.723Mixed
apply STKIM masking at training, versus MHIM-MIL teacher-student maskingABMIL · CAMELYON16 (ResNet-18)training time per epoch: 20.8s -> 8.0sImproved
add STKIMABMIL attention · CAMELYON16top-10 cumulative attention: 0.87 -> 0.6Improved

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