Research map / Video / Video-language understanding
CVA: Context-aware Video-text Alignment for Video Temporal Grounding
Sungho Moon, Seunghun Lee, Jiwan Seo, Sunghoon Im
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add query-aware context diversification (QCD) (ablation) | baseline · QVHighlights val | R1@0.7: 46.77 -> 51.98 | Improved |
| use QCD instead of query-agnostic mix (ablation) | baseline · QVHighlights val | R1@0.7: 48.39 -> 51.98 | Improved |
| add CTE on top of QCD (ablation) | QCD model · QVHighlights val | R1@0.7: 51.98 -> 52.63 | Improved |
| add CBD loss on top of QCD (ablation) | QCD model · QVHighlights val | R1@0.7: 51.98 -> 53.02 | Improved |
| add all three components (ablation) | baseline · QVHighlights val | R1@0.7: 46.77 -> 54.84 | Improved |
| add semantically hard negatives to CBD (Nhard 0 -> 2) (ablation) | CBD · QVHighlights val | R1@0.7: 51.94 -> 54.89 | Improved |
| increase Nhard from 5 to 10 (ablation) | CBD · QVHighlights val | R1@0.7: 54.84 -> 52.71 | Hurt |
| temporal-only negatives in CBD (Nhard=0) (ablation) | CBD · QVHighlights val | R1@0.7: 51.98 -> 51.94 | No clear effect |
| raise upper percentile to 90 (ablation) | QCD · QVHighlights val | R1@0.7: 50.92 -> 48.88 | Hurt |
| raise lower percentile to 40 (ablation) | QCD · QVHighlights val | R1@0.7: 50.92 -> 49.71 | Hurt |
| add learnable queries only to CTE (ablation) | QCD model · QVHighlights val | R1@0.7: 51.98 -> 52.33 | No clear effect |
| add windowed self-attention only to CTE (ablation) | QCD model · QVHighlights val | R1@0.7: 51.98 -> 52.13 | No clear effect |
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.
- BAM-DETR — BAM-DETR: Boundary-Aligned Moment Detection Transformer for Temporal Sentence Grounding in Videos (ECCV 2024) · better on 3 of 3 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
- Video-Context Aligned Transformer for Video Question Answering (AAAI 2024 · similarity 0.89)
- VidLA: Video-Language Alignment at Scale (CVPR 2024 · similarity 0.89)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.88)
- VideoComp: Advancing Fine-Grained Compositional and Temporal Alignment in Video-Text Models (CVPR 2025 · similarity 0.88)
- TOPA: Extending Large Language Models for Video Understanding via Text-Only Pre-Alignment (NeurIPS 2024 · similarity 0.88)
- Mitigating Semantic Collapse in Partially Relevant Video Retrieval (NeurIPS 2025 · similarity 0.88)
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.
