AI Paper Atlas

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

Transitive Representation Learning Enhances Histopathology Annotation

Moritz Schaefer, Zoe Piran, Nils Philipp Walter, Animesh Awasthi, Christoph Bock, Jure Leskovec, Zinaida Good

ICML 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Medical vision-language models

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
train on I-G and G-T pairs with transitive alignment instead of bimodal I-T pretrainingSpatialWhisperer · PathoCell, Lizardmean AUROC: PathoCell 0.566 -> 0.621; Lizard 0.642 -> 0.749Improved
train on I-G and G-T pairs instead of bimodal I-T QUILT-1M trainingSpatialWhisperer · PanNuke cell typemean AUROC: 0.713 -> 0.687Hurt
joint trimodal alignment instead of two-stage image->expression->cell-type pipelineSpatialWhisperer · PathoCell, Lizard, PanNukemean AUROC: 0.550 -> 0.621; 0.662 -> 0.749; 0.599 -> 0.687Improved
joint alignment of image, expression and annotation instead of marker-gene-only bridgeSpatialWhisperer vs OmiCLIP · PathoCellmean AUROC: 0.491 -> 0.621Improved
harmonize text label style of I-T data with G-T data (LLM rewrite) (ablation)QUILT-1M labels · QUILT-1M image-text retrievalAUROC: 0.645 -> 0.695Improved
use curated instead of uncurated QUILT-1M data in trimodal training (ablation)SpatialWhisperer · PathoCellmean AUROC: 0.609 -> 0.645Improved
add transitive data (trimodal) to subsampled task-matched data (ablation)bimodal training · low-data regimes (G->T, I<->G retrieval, skin conditions)AUROCImproved
add transitive data at full task-matched data (ablation)bimodal training · full dataAUROCNo clear effect
add I-T supervision to T-G + G-I bridge training (ablation)SpatialWhisperer · PanNuke 19-class tissue-of-originAUROC: 0.586 -> 0.649Improved

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