AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

TaxoMIL: Taxonomy-Constrained Learning for Hierarchical Whole Slide Image Analysis

Chaeyeon Lee, Khang Quoc, Jinsol Song, Yosep Chong, Kwangil Yim, JIN TAE KWAK

ECCV 2026 · Image generation › Text-to-image diffusion · approach: Text generation and structured prediction · 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 cyclic loss scheduling to HIER, ITAL and CL (ablation)TaxoMIL · GastWSI, BRACS, PANDAholistic accuracy (%): GastWSI 63.07 -> 64.55; BRACS 72.22 -> 75.78; PANDA 51.41 -> 54.60Improved
add HIER and ITAL losses on top of GEN+CL (ablation)TaxoMIL · BRACSholistic accuracy (%): 62.96 -> 72.22Improved
add HIER and ITAL losses on top of GEN+CL (no cyclic) (ablation)TaxoMIL · PANDAholistic accuracy (%): 51.97 -> 51.41No clear effect
add supervised contrastive loss to GEN (ablation)TaxoMIL · BRACSholistic accuracy (%): 66.67 -> 62.96Hurt
add supervised contrastive loss to GEN (ablation)TaxoMIL · GastWSI, PANDAholistic accuracy (%): GastWSI 61.58 -> 62.60; PANDA 49.53 -> 51.97Improved
add label-context token e_ctx to decoder prefix (ablation)TaxoMIL · GastWSI, BRACS, PANDAholistic accuracy (%): GastWSI 62.51 -> 64.55; BRACS 70.37 -> 75.78; PANDA 50.56 -> 54.60Improved
add learnable image token <|img|> to decoder prefix (ablation)TaxoMIL · GastWSI, BRACS, PANDAholistic accuracy (%): GastWSI 62.05 -> 64.55; BRACS 70.37 -> 75.78; PANDA 53.10 -> 54.60Improved
use autoregressive text decoder instead of classification heads (ablation)TaxoMIL · GastWSI, BRACS, PANDAholistic accuracy (%): GastWSI 63.81 -> 64.55; BRACS 72.22 -> 75.78; PANDA 52.35 -> 54.60Improved
use GPT-2 small instead of GPT-2 medium decoder (ablation)TaxoMIL · BRACSholistic accuracy (%): 66.67 -> 75.78Improved
TaxoMIL vs strongest baseline per dataset (holistic)TaxoMIL · GastWSI, BRACS, PANDAholistic accuracy (%): GastWSI 63.26 -> 64.55; BRACS 68.52 -> 75.78; PANDA 50.28 -> 54.60Improved
flat MIL classifiers vs hierarchy-aware MIL (HMIL, Chang et al., HiClass)hierarchical MIL baselines · PANDA, GastWSIholistic accuracy (%): GastWSI HMIL 55.07 vs ABMIL 61.02No clear effect

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