Research map / Video / Video-language understanding
Explore Inter-contrast between Videos via Composition for Weakly Supervised Temporal Sentence Grounding
Jiaming Chen, Weixin Luo, Wei Zhang, Lin Ma
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 |
|---|---|---|---|
| Train with two relevant queries rather than QueryA only (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.1: 67.72 vs 71.86 | Improved |
| Train with two relevant queries rather than QueryA only (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.3: 44.60 vs 46.62 | Improved |
| Train with two relevant queries rather than QueryA only (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.5: 27.24 vs 29.52 | Improved |
| Add negative-query training (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.1: 68.42 vs 71.86 | Improved |
| Add negative-query training (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.3: 44.62 vs 46.62 | Improved |
| Add negative-query training (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.5: 28.76 vs 29.52 | Improved |
| Add composition ratio prediction (CRP) (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.1: 71.52 vs 71.86 | Improved |
| Add composition ratio prediction (CRP) (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.3: 45.98 vs 46.62 | Improved |
| Add composition ratio prediction (CRP) (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.5: 29.01 vs 29.52 | Improved |
| Add masked word prediction (MWP) (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.1: 69.12 vs 71.86 | Improved |
| Add masked word prediction (MWP) (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.3: 44.31 vs 46.62 | Improved |
| Add masked word prediction (MWP) (ablation) | Video composition grounding model · ActivityNet-Captions | R@1, IoU=0.5: 27.89 vs 29.52 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- VLANet — VLANet: Video-Language Alignment Network for Weakly-Supervised Video Moment Retrieval (ECCV 2020) · better on 0 of 1 shared result
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
- Weakly Supervised Temporal Sentence Grounding With Gaussian-Based Contrastive Proposal Learning (CVPR 2022 · similarity 0.92)
- Multi-Pair Temporal Sentence Grounding via Multi-Thread Knowledge Transfer Network (AAAI 2025 · similarity 0.91)
- Siamese Learning with Joint Alignment and Regression for Weakly-Supervised Video Paragraph Grounding (CVPR 2024 · similarity 0.90)
- Learning Multi-Scale Video-Text Correspondence for Weakly Supervised Temporal Article Gronding (AAAI 2024 · similarity 0.90)
- Empower Words: DualGround for Structured Phrase and Sentence-Level Temporal Grounding (NeurIPS 2025 · similarity 0.89)
- Universal Video Temporal Grounding with Generative Multi-modal Large Language Models (NeurIPS 2025 · similarity 0.89)
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.
