Research map / Video / Video-language understanding
StoryVideoQA: Scaling Deep Video Understanding with a Large-Scale, Multi-Genre and Auto-Generated Dataset
Zhengqian Wu, Zhixian Liu, Aodong Chen, Jingyang Zhang, Ruizhe Li, Hanlin Ge, Zhongyuan Wang, Chunxia Xiao, Chao Liang
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 a supervisor agent that deletes erroneous QAs and gives feedback (ablation) | StoryMind generator · QA generation (2,000 QAs) | accuracy (%): 49.50 -> 62.30 | Improved |
| add a supervisor agent (ablation) | StoryMind generator · QA generation (2,000 QAs) | time per QA (min): 0.081 -> 0.365 | Hurt |
| use majority answer voting instead of unanimous consistency (ablation) | QA filtration · QA filtration quality | recall (%): 54.15 -> 67.02 | Improved |
| add correctness judgement (ablation) | QA filtration · StoryVideoQA-R | accuracy (%): 80.71 -> 86.16 | Improved |
| add answer voting after correctness judgement (ablation) | QA filtration · StoryVideoQA-R | accuracy (%): 86.16 -> 90.12 | Improved |
| add character identification (ablation) | PlotTree · StoryVideoQA-G | average accuracy (%): 82.37 -> 86.50 | Improved |
| add character identification (ablation) | Video2RAG · StoryVideoQA-G | average accuracy (%): 80.22 -> 83.63 | Improved |
| add Ins. module (ablation) | PlotTree · StoryVideoQA-G | average accuracy (%): 80.22 -> 83.57 | Improved |
| add Dia. module (ablation) | PlotTree · StoryVideoQA-G | average accuracy (%): 80.22 -> 83.79 | Improved |
| add Tree module (ablation) | PlotTree · StoryVideoQA-G | average accuracy (%): 80.22 -> 82.37 | Improved |
| add Tree module to Ins.+Dia. (ablation) | PlotTree · StoryVideoQA-G | average accuracy (%): 83.63 -> 86.50 | Improved |
| add Dia. module to Ins.+Tree (ablation) | PlotTree · StoryVideoQA-G | average accuracy (%): 86.00 -> 86.50 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- VideoTree — VideoTree: Adaptive Tree-based Video Representation for LLM Reasoning on Long Videos (CVPR 2025) · better on 2 of 2 shared results
- VideoChat-Flash — VideoChat-Flash: Hierarchical Compression for Long-Context Video Modeling (ICLR 2026) · 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
- FriendsQA: A New Large-Scale Deep Video Understanding Dataset with Fine-grained Topic Categorization for Story Videos (AAAI 2025 · similarity 0.96)
- MovieQA: Understanding Stories in Movies Through Question-Answering (CVPR 2016 · similarity 0.92)
- VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection (CVPR 2025 · similarity 0.87)
- VTQA: Visual Text Question Answering via Entity Alignment and Cross-Media Reasoning (CVPR 2024 · similarity 0.87)
- LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding (NeurIPS 2024 · similarity 0.87)
- Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA (NeurIPS 2026 · 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.
