Research map / Video / Video-language understanding
Semantic Conditioned Dynamic Modulation for Temporal Sentence Grounding in Videos
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 sentence-conditioned dynamic modulation instead of plain batch normalization (ablation) | SCDM temporal convolution model · Charades-STA | R@5, IoU@0.7: 49.35 -> 58.08 | Helped |
| Add sentence-conditioned dynamic modulation instead of global-sentence FC fusion (ablation) | SCDM temporal convolution model · Charades-STA | R@5, IoU@0.7: 49.81 -> 58.08 | Helped |
| Add sentence-conditioned dynamic modulation instead of global-sentence element-wise multiplication (ablation) | SCDM temporal convolution model · Charades-STA | R@5, IoU@0.7: 51.02 -> 58.08 | Helped |
| Dynamically vary modulation vectors by temporal feature unit rather than using global, fixed vectors (ablation) | SCDM temporal convolution model · Charades-STA | R@5, IoU@0.7: 54.57 -> 58.08 | Helped |
| Use Ours-SCDM rather than MAN | Charades-STA grounding · Charades-STA | R@5, IoU@0.5: 86.23 -> 74.43 | Hurt |
| Use Ours-SCDM rather than ACL | TACoS grounding · TACoS | R@5, IoU@0.3: 42.15 -> 40.16 | Hurt |
Closest papers by meaning
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.90)
- Multi-Pair Temporal Sentence Grounding via Multi-Thread Knowledge Transfer Network (AAAI 2025 · similarity 0.89)
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · similarity 0.89)
- Temporal Sentence Grounding with Relevance Feedback in Videos (NeurIPS 2024 · similarity 0.88)
- Learning Multi-Scale Video-Text Correspondence for Weakly Supervised Temporal Article Gronding (AAAI 2024 · similarity 0.87)
- Siamese Learning with Joint Alignment and Regression for Weakly-Supervised Video Paragraph Grounding (CVPR 2024 · similarity 0.87)
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.