AI Paper Atlas

Research map / Video / Video-language understanding

GroundVTS: Visual Token Sampling in Multimodal Large Language Models for Video Temporal Grounding

Rong Fan, Kaiyan Xiao, Minghao Zhu, Liuyi Wang, Kai Dai, Zhao Yang

CVPR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 0 citations (OpenAlex)

Read the paper

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 query-guided visual token sampling (VTS) on top of grounding fine-tuned modelQwen2.5VL-7B-G · Charades-STA testmIoU: 31.7 -> 50.1Improved
add VTS to grounding fine-tuned modelInternVL3.5-8B-G · Charades-STA testmIoU: 39.4 -> 41.6Improved
add VTSInternVL3.5-8B-G · QVHighlights val (HD)Hit@1: 39.8 -> 88.4Improved
add VTSQwen2.5VL-7B-G · QVHighlights val (HD)mAP: 34.4 -> 35.7Improved
use token-level instead of frame-level query selection (ablation)GroundVTS-Q · Charades-STA testmIoU: 41.6 -> 50.1Improved
use query-guided token sampling instead of uniform (ablation)GroundVTS-Q · Charades-STA testmIoU: 29.3 -> 50.1Improved
use random token dropping instead of uniform frame sampling (ablation)GroundVTS-Q · Charades-STA testmIoU: 29.3 -> 35.7Improved
keep original positional encodings for selected tokens (ablation)GroundVTS-Q · Charades-STA testmIoU: 9.5 -> 50.1Improved
train VTS before use (random-initialized VTS) (ablation)GroundVTS-Q · Charades-STA testmIoU: 5.6 -> 20.9Improved
add stage 2 joint LoRA adaptation (ablation)GroundVTS-Q · Charades-STA testmIoU: 32.4 -> 50.1Improved
add stage 1 VTS warm-up (ablation)GroundVTS-Q · Charades-STA testmIoU: 47.4 -> 50.1Improved
increase frame rate beyond 2.4 FPS (ablation)Qwen2.5VL-7B (pretrained) · Charades-STA testmIoU: 47.8 peak at 2.0-2.4 FPS, then sharp dropHurt

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