AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

CAR-MIL: Counterfactual Attention Regularization for Multiple Instance Learning

Imane Chraki, Pierre Marza, Stergios Christodoulidis, Maria Vakalopoulou

ECCV 2026 · Architectures and efficiency › Attention and transformers · approach: Sparse and structured attention · 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
add counterfactual attention regularization (L1)ABMIL · MNIST-bags Adjacent Pairsinstance AUPRC+/-: 61.5 ± 0.9 -> 84.6 ± 5.2Improved
add counterfactual attention regularization (L1)ABMIL · MNIST-bags Four Bagsinstance AUPRC+/-: 53.1 ± 0.1 -> 89.7 ± 0.9Improved
add counterfactual attention regularization (cosine)ABMIL · TCGA-LUAD (TP53)AUC: 74.0±4.3 -> 78.1±4.1Improved
add counterfactual attention regularization (cosine)ABMIL · BRACSbalanced accuracy: 38.2±4.1 -> 43.1±2.0Improved
add counterfactual attention regularizationABMIL · TCGA-BRCAAUC: 95.3±1.7 -> 95.5±2.2No clear effect
add counterfactual attention regularization to DSMIL (ablation)DSMIL · TCGA-LUAD (TP53)AUC: 66.9±4.7 -> 71.2±4.1Improved
add counterfactual attention regularization to TransMIL (ablation)TransMIL · TCGA-LUAD (TP53)AUC: 71.7±5.4 -> 73.6±4.4Improved
add counterfactual attention regularization to TransMIL (ablation)TransMIL · BRACSbalanced accuracy: 40.0±3.6 -> 42.6±6.1Improved
add counterfactual attention regularization to TransMIL (ablation)TransMIL · CAMELYON16AUPRC+: 28.2±3.6 -> 32.6±0.3Improved
add counterfactual attention regularization to DSMIL (ablation)DSMIL · CAMELYON16AUPRC+: 80.6±11.2 -> 87.8±2.9Improved
use cosine instead of L1 attention-logit distance (ablation)CAR-MIL · LUAD, BRACS, BRCA, NSCLC, CAMELYON16AUC/BACC: LUAD AUC 76.4±4.9 -> 78.1±4.1; CAMELYON16 AUPRC+ 94.8±1.0 -> 91.7±0.6Mixed
add the evidence-differential and proximity losses (alpha, lambda in 0.2, 0.8, 1.0) (ablation)ABMIL · TCGA-BRCAAUC: 94.5 -> 95.3 to 96.0Improved

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