AI Paper Atlas

Research map / Video / Video-language understanding

Sparse-Dense Side-Tuner for efficient Video Temporal Grounding

David Pujol-Perich, Sergio Escalera, Albert Clapés

ICCV 2025 · 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
reuse frozen AdaptivePool instead of CLS pooling for InternVideo2 features (ablation)SDST · QVHighlights valMR mAP avg: 50.53 -> 55.60Improved
reuse frozen AdaptivePool instead of average pooling (ablation)SDST · QVHighlights valMR mAP avg: 53.44 -> 55.60Improved
sample intermediate layers from shallower depth ([10,20,30,40] vs [37,38,39,40]) (ablation)SDST · QVHighlights valMR mAP avg: 55.60 -> 52.47Hurt
use intermediate-layer features instead of last-layer features for refinement steps (ablation)SDST · QVHighlights valMR mAPMixed
replace deformable cross-attention with RDSA (ablation)SDST · QVHighlights valMR R1@0.7 and mAP: 58.19 -> 60.90 R1@0.7; 54.27 -> 55.60 mAPImproved
replace standard cross-attention with RDSA (ablation)SDST · QVHighlights valMR mAP avg: 42.72 -> 55.60Improved
add MixedInit decoder query initialization to deformable CA (ablation)Def. CA · QVHighlights valMR mAP avg: 54.27 -> 54.94No clear effect
add PureInit decoder query initialization to deformable CA (ablation)Def. CA · QVHighlights valMR mAP avg: 54.27 -> 52.92Hurt
condition sparse stream on post CA+SA+FFN output instead of raw video representation (ablation)SDST · QVHighlights valMR mAP avg: 50.74 -> 55.60Improved
add cross-attention (CA) module to sparse stream (ablation)SDST · QVHighlights valMR mAP avg: 54.47 -> 55.60Improved
add self-attention (SA) module to sparse stream (ablation)SDST · QVHighlights valMR mAP avg: 54.32 -> 55.60Improved
add RDSA module to sparse stream (ablation)SDST · QVHighlights valMR mAP avg: 29.09 -> 55.60Improved

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