AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Comprehensive language–image pre-training for 3D medical image understanding

Tassilo Wald, Ibrahim Ethem Hamamci, Yuan Gao, Sam Bond-Taylor, Harshita Sharma, Maximilian Ilse, Cynthia Lo, Olesya Melnichenko, Anton Schwaighofer, Noel Codella, Maria Teodora Wetscherek, Klaus Maier-Hein and 4 more

ECCV 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Adapting VLMs for zero-shot classification · 28 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 opposite sentence loss (ablation)COLIPRI-CRM · Zero-shot short prompts, CT-RATE valAUROC: 74.09 -> 81.66Improved
add report generation objectiveCOLIPRI-C · CT-RATE classification probeAUPRC: 55.13 -> 57.44Improved
add MAE objectiveCOLIPRI-C · CT-RATE classification probeAUPRC: 55.13 -> 61.52Improved
add report generation on top of MAECOLIPRI-CM · CT-RATE retrievalRecall@5: 34.16 -> 35.10Improved
add report generation on top of MAECOLIPRI-CM · CT-RATE zero-shotAUROC: 80.52 -> 79.81No clear effect
add MAE objectiveCOLIPRI-C · LiTS segmentation (37.5k steps)DSC: 76.77 -> 79.97Improved
pre-train with COLIPRI-CRM instead of training from scratchPrimus-M · Lung segmentation (250k steps)DSC: 69.11 -> 67.56Hurt
add sentence shuffle augmentation (ablation)default CLIP · CT-RATE val retrievalRecall@5: 22.64 -> 28.57Improved
add short sentence augmentation p=0.10 (ablation)default CLIP · CT-RATE val short zero-shotAUROC: 66.91 -> 71.15Improved
train on impressions instead of findings (ablation)default CLIP · CT-RATE val short zero-shotAUROC: 66.91 -> 62.40Hurt
re-translate reports with GPT-4o (ablation)default CLIP · CT-RATE val short zero-shotAUROC: 66.91 -> 60.11Hurt
use BiomedCLIP text encoder instead of CXR-BERT (ablation)default CLIP · CT-RATE val retrievalRecall@5: 22.64 -> 13.28Hurt

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