Research map / Video / Video-language understanding
Video Question Answering with Procedural Programs
Rohan Choudhury, Koichiro Niinuma, Kris Kitani, Laszlo A Jeni
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 video_query module (frame-voting video QA) on top of single-image querying (ablation) | ProViQ · ActivityNet-QA | accuracy: 27.7 -> 35.0 | Improved |
| add video_query module on top of single-image querying (ablation) | ProViQ · NeXT-QA | accuracy: 49.3 -> 53.2 | Improved |
| add filter_property and filter_object modules | ProViQ · long video datasets (ActivityNet, NeXT-QA) | accuracy | Improved |
| add the get_caption module (ablation) | ProViQ · NeXT-QA | accuracy: 55.9 -> 63.8 | Improved |
| add more modules to the API (ablation) | ProViQ · individual datasets (e.g. MSVD, MSRVTT) | accuracy: MSVD 37.5 -> 22.8 at one stage | Mixed |
| add in-context example programs to the prompt (0 to 1 and more) | ProViQ · video QA benchmarks | accuracy | Improved |
| use video-centric modules instead of image-centric ViperGPT API | visual programming · iVQA | accuracy: 41.4 -> 50.7 | Improved |
| add get_summary long video module with LaViLa captions | ProViQ · EgoSchema | accuracy: 26.1 -> 57.1 | Improved |
| apply ProViQ to datasets with ambiguous labels (MSR-VTT, MSVD) | ProViQ · MSVD-QA, MSRVTT-QA | accuracy: MSVD 37.5 vs supervised 54.8 | No clear effect |
Closest papers by meaning
- Object-centric Video Question Answering with Visual Grounding and Referring (ICCV 2025 · similarity 0.88)
- MoReVQA: Exploring Modular Reasoning Models for Video Question Answering (CVPR 2024 · similarity 0.87)
- VTQA: Visual Text Question Answering via Entity Alignment and Cross-Media Reasoning (CVPR 2024 · similarity 0.87)
- Track the Answer: Extending TextVQA from Image to Video with Spatio-Temporal Clues (AAAI 2025 · similarity 0.87)
- Language-aware Visual Semantic Distillation for Video Question Answering (CVPR 2024 · similarity 0.86)
- When Open-Vocabulary Visual Question Answering Meets Causal Adapter: Benchmark and Approach (AAAI 2025 · similarity 0.86)
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.
