AI Paper Atlas

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

Differentiable Zooming for Multiple Instance Learning on Whole-Slide Images

Kevin Thandiackal, Boqi Chen, Pushpak Pati, Guillaume Jaume, Drew F. K. Williamson, Maria Gabrani, Orcun Goksel

ECCV 2022 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 65 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
use differentiable Top-K instead of random selection at lowest magnification (ablation)ZoomMIL-Eff · BRIGHT (ZoomMIL-Eff)weighted F1: 61.0 -> 68.1Improved
use differentiable Top-K instead of random 4K selection at 2.5x (ablation)ZoomMIL-Eff · BRIGHT (ZoomMIL-Eff)weighted F1: 59.6 -> 68.1Improved
use differentiable Top-K instead of non-differentiable Top-K (ablation)ZoomMIL-Eff · BRIGHT (ZoomMIL-Eff)weighted F1: 59.9 -> 68.1Improved
use dual gated attention instead of single gated attention (ablation)ZoomMIL-Eff · BRIGHT (ZoomMIL-Eff)weighted F1: 59.6 -> 68.1Improved
add lowest-magnification features to highest-magnification features (ablation)ZoomMIL-Eff · BRIGHT (ZoomMIL-Eff)weighted F1: 62.7 -> 68.1Improved
use sum-pooling instead of concatenation of magnification features (ablation)ZoomMIL-Eff · BRIGHT (ZoomMIL-Eff)weighted F1: 64.9 -> 68.1Improved
add third magnificationZoomMIL · CRCweighted F1: 90.3 -> 92.0Improved
add third magnificationZoomMIL · BRIGHTweighted F1: 66.0 -> 68.3Improved
ZoomMIL vs CLAM-SBZoomMIL · BRIGHTweighted F1: 63.1 -> 68.3Improved
ZoomMIL vs TransMIL with lowest magnification 10xZoomMIL · CAMELYON16accuracy: 85.3 -> 84.2Hurt
zooming instead of single-magnification CLAM-SBZoomMIL · BRIGHTTFLOPs: 16.45 -> 1.29Improved
large vs small metastatic regionsZoomMIL · CAMELYON16 validationweighted F1: 86.7 -> 96.2Mixed

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