AI Paper Atlas

Research map / Video / Video-language understanding

Decoupling Moment from Event for Video Temporal Grounding

Yuda Zou, Boxiang Zhou, Xin Zhou, YIBO CHEN, Dejia Song, Xu Tang, Yao Hu, Yongchao Xu

ECCV 2026 · Video › Video-language understanding · approach: Temporal video grounding · 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
add moment-event synergistic supervision (moment queries) to one-to-one matching (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 52.83 -> 55.41Improved
use one-to-many matching where all positives regress the full GT span (ablation)standard one-to-one DETR · QVHighlights val (InternVideo)Avg mAP: 52.83 -> 53.35Improved
replace one-to-many with moment-query supervision (ablation)one-to-many variant · QVHighlights val (InternVideo)Avg mAP: 53.35 -> 55.41Improved
lower moment confidence threshold tau_cls from 0.3 to 0.2 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 55.41 -> 55.10No clear effect
raise moment confidence threshold tau_cls from 0.3 to 0.4 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 55.41 -> 55.09No clear effect
lower containment threshold tau_in from 0.7 to 0.5 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 55.41 -> 55.09No clear effect
raise containment threshold tau_in from 0.7 to 0.8 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 55.41 -> 55.16No clear effect
mine K=1 moment query per event instead of 2 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 55.41 -> 55.08Hurt
mine K=5 moment queries per event instead of 2 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 55.41 -> 54.86Hurt
mine K=2 moment queries per event instead of none (K=0) (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 52.83 -> 55.41Improved
raise moment loss weight lambda_m from 0.2 to 1.0 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 55.41 -> 54.47Hurt
add moment loss with lambda_m=0.1 instead of 0.0 (ablation)ME-DETR · QVHighlights val (InternVideo)Avg mAP: 52.83 -> 55.14Improved

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