Research map / Video / Video-language understanding
Enhancing Video-Language Representations With Structural Spatio-Temporal Alignment
Hao Fei, Shengqiong Wu, Meishan Zhang, Min Zhang, Tat‐Seng Chua, Shuicheng Yan
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 Finsta structural scene-graph alignment | InternVideo · K400 | Top-1 accuracy: 91.1 -> 93.7 | Improved |
| add Finsta | HDVILA · MSR-VTT captioning | METEOR: 32.4 -> 36.9 | Improved |
| add Finsta | HDVILA · LSMDC retrieval | R@1: 17.4 -> 25.3 | Improved |
| add Finsta | Video-LLaVA · MSVD-QA | accuracy: 65.7 -> 72.5 | Improved |
| add Finsta | All-in-one · MSR-VTT captioning | METEOR: 32.7 -> 34.2 | Improved |
| add Finsta | LFVILA (L-Vid) · ActivityNet video-paragraph retrieval | R@1: 35.3 -> 69.8 | Improved |
| add Finsta | LFVILA (L-Vid) · How2QA | accuracy: 76.1 -> 84.8 | Improved |
| add Finsta in zero-shot instead of fine-tuned setting | Video-LLaVA · MSVD-QA zero-shot | accuracy: 25.1 -> 45.3 | Improved |
| add Finsta in zero-shot | InternVideo · K400 zero-shot | Top-1 accuracy: 53.3 -> 68.9 | Improved |
| add temporal coreference edges in the dynamic scene graph (ablation) | Finsta · 6-task average | average score: 59.2 -> 62.9 | Improved |
| add adverbial modifier nodes in the textual scene graph (ablation) | Finsta · 6-task average | average score: 60.0 -> 62.9 | Improved |
| add cross-modal coreference edges in the holistic scene graph (ablation) | Finsta · 6-task average | average score: 60.5 -> 62.9 | Improved |
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.
- VideoMAE — VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training (NeurIPS 2022) · better on 1 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
- Structured Video-Language Modeling with Temporal Grouping and Spatial Grounding (ICLR 2024 · similarity 0.90)
- LLaVA-ST: A Multimodal Large Language Model for Fine-Grained Spatial-Temporal Understanding (CVPR 2025 · similarity 0.89)
- Video-Panda: Parameter-efficient Alignment for Encoder-free Video-Language Models (CVPR 2025 · similarity 0.89)
- VidLA: Video-Language Alignment at Scale (CVPR 2024 · similarity 0.89)
- TOPA: Extending Large Language Models for Video Understanding via Text-Only Pre-Alignment (NeurIPS 2024 · similarity 0.88)
- V-CAST: Video Curvature-Aware Spatio-Temporal Pruning for Efficient Video Large Language Models (NeurIPS 2026 · 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.
