AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

MULTIMODALITY AS SUPERVISION: SELF-SUPERVISED SPECIALIZATION TO THE TEST ENVIRONMENT VIA MULTIMODALITY

Kunal Pratap Singh, Ali Garjani, Rishubh Singh, Muhammad Uzair Khattak, Jason Toskov, Efe Tarhan, Andrei Atanov, Oğuzhan Fatih Kar, Amir Zamir

ICLR 2026 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervision in supervised pipelines · 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
pre-train on test-space multimodal data with TST-MM instead of 4M-21 generalist4M-21 · ProcTHOR semantic segmentationmIoU: 53.24 -> 60.85Improved
fine-tune 4M-21 on test-space data (adapted)TST-MM · Scannet++ object detectionmAP: 31.54 -> 35.83Improved
adapt pretrained 4M-21 to test space4M-21 · Scannet++ semantic segmentationmIoU: 27.59 -> 36.44Improved
add semantic segmentation modality (ablation)TST-MM · ProcTHOR semantic segmentationmIoU: 59.43 -> 60.85Improved
add object detection modality (ablation)TST-MM · ProcTHOR semantic segmentationmIoU: 59.38 -> 60.85Improved
add semantic segmentation, detection and SAM edge modalities (ablation)TST-MM · ProcTHOR semantic segmentationmIoU: 55.39 -> 60.85Improved
add semantic segmentation, detection and SAM edge modalities (ablation)TST-MM · ProcTHOR object detectionmAP: 45.97 -> 49.38Improved
add object detection modality (ablation)TST-MM · ProcTHOR object detectionmAP: 49.34 -> 49.38No clear effect
add SAM edges modality to RGB alone (ablation)TST-MM · Scannet++ semantic segmentationmIoU: +7.8%Improved
add SAM edges modality to the seven others (ablation)TST-MM · Scannet++ semantic segmentationmIoU: 1.5%Improved
use untrained pseudolabel networks as modalities (ablation)TST-MM · test-space segmentationmIoUHurt
mix transfer-set RGB frames into test-space pre-training (ablation)TST-MM (ViT-S) · test-space segmentationmIoUImproved

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