AI Paper Atlas

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

T-REN: Learning Text-Aligned Region Tokens Improves Dense Vision-Language Alignment and Scalability

Savya Khosla, Sethuraman T V, Aryan Chadha, Alex Schwing, Derek Hoiem

ECCV 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Recaptioning and text-supervised pretraining · 2 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 T-REN region tokens on top of dino.txtDINOv3 dino.txt · ADE20K OVSSmIoU: 24.7 -> 30.6Improved
add T-REN region tokensDINOv3 dino.txt · Cityscapes OVSSmIoU: 36.9 -> 52.7Improved
add T-REN region tokensDINOv3 dino.txt · Visual Haystacks D=1Kaccuracy (%): 52.1 -> 65.2Improved
add T-REN track tokensDINOv3 dino.txt · Ego4D query localizationRecall@1: 36.8 -> 52.4Improved
add T-REN track tokensDINOv3 dino.txt · VSPW scene parsingmIoU: 20.7 -> 38.3Improved
add region pooling jointly with text alignment (ablation)T-REN · ADE20K OVSSmIoU: 25.4 -> 30.6Improved
add joint text alignment training to region pooling (ablation)T-REN · ADE20K OVSSmIoU: 19.5 -> 30.6Improved
add region pooling to patch-level text alignment (ablation)T-REN · Visual Haystacks retrieval D=10accuracy (%): 76.1 -> 87.2Improved
add token merging within and across frames (ablation)T-REN · VSPW scene parsingmIoUNo clear effect
add temporal token merging (tau_track 0.5) (ablation)T-REN · Ego4D query localization subsetrecall tIoU=0.25 (%): 39.4 -> 57.7Improved
raise tau_track to 0.9 (ablation)T-REN · Ego4D query localization subsetrecall tIoU=0.25 (%): 57.7 -> 45.2Hurt
predict k=3 region tokens per prompt instead of one (ablation)T-REN · COCO retrievalrecallImproved

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