AI Paper Atlas

Research map / Video / Video-language understanding

Decoupled Seg Tokens Make Stronger Reasoning Video Segmenter and Grounder

Jisheng Dang, Xudong Wu, Haowen Yan, Huicheng Zheng, Wei‐Shi Zheng, Jianhuang Lai, Bin Hu, Tat‐Seng Chua

TPAMI 2026 · Video › Video-language understanding · approach: Long video understanding · 5 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 text understanding pre-training of SAM-2 (ablation)DeSa2VA-4B · RefCOCOcIoU: 79.2 -> 82.6Improved
add text understanding pre-training of SAM-2 (ablation)DeSa2VA-4B · MeViSJ&F: 45.9 -> 47.2Improved
add text pre-training to the Sa2VA baseline without decoupling (ablation)Sa2VA-4B · RefCOCOcIoU: 78.9 -> 73.4Hurt
add text pre-training to the Sa2VA baseline without decoupling (ablation)Sa2VA-1B · RefCOCOcIoU: 77.4 -> 72.8Hurt
add text pre-training with decoupling (ablation)DeSa2VA-1B · MeViSJ&F: 40.9 -> 45.1Improved
add mask reprompting (dense mask prompt) (ablation)DeSa2VA-4B · RefCOCOcIoU: 81.3 -> 82.6Improved
add mask reprompting (dense mask prompt) (ablation)Sa2VA-4B · RefCOCOcIoU: 78.9 -> 80.9Improved
add mask reprompting (ablation)DeSa2VA-4B · SEED-Benchaccuracy: 73.2 -> 73.3No clear effect
add the decoupling module (text/visual linear projections) to Sa2VADeSa2VA-4B vs Sa2VA-4B · RefCOCO+cIoU: 71.7 -> 77.8Improved
add decoupling to Sa2VA on video QA and image QADeSa2VA-4B · SEED-Benchaccuracy: 73.3 -> 73.3No clear effect
add image segmentation data to training (ablation)DeSa2VA-4B · RefCOCOcIoU: 21.5 -> 82.6Improved
add video segmentation data to training (ablation)DeSa2VA-4B · RefCOCOgcIoU: 76.8 -> 79.2Mixed

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