Research map / Video / Video-language understanding
From Representation to Reasoning: Towards Both Evidence and Commonsense Reasoning for Video Question-Answering
Check 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 |
|---|---|---|---|
| Use video features rather than question-answer text alone | EVQA versus BlindQA · Causal-VidQA | overall accuracy: 22.43 -> 27.52 | Helped |
| Add spatio-temporal attention and memory fusion | CoMem versus EVQA · Causal-VidQA | overall accuracy: 27.52 -> 40.51 | Helped |
| Use BERT instead of GloVe for B2A text representation | B2A · Causal-VidQA | overall accuracy: 43.26 -> 49.11 | Helped |
| Use BERT instead of GloVe for B2A text representation | B2A · Causal-VidQA counterfactual questions | counterfactual Q → AR accuracy: 25.29 -> 35.16 | Helped |
| Use graph-based cross-modal interaction | HGA and B2A · Causal-VidQA | overall accuracy | Helped |
| Require selecting both the answer and reason correctly | Causal-VidQA predictive and counterfactual evaluation · Causal-VidQA | accuracy: HCRN prediction Q → A: 51.74; Q → AR: 32.57; B2A counterfactual Q → A: 53.27; Q → AR: 35.16 | Hurt |
| Choose candidates using answer length or semantic similarity heuristics | Causal-VidQA diagnosis baselines · Causal-VidQA test | overall accuracy: 11.84; 11.98; 11.50; 12.24; 11.88; 12.03 | No clear effect |
| Replace the segmentation label [bed_1] with the word bed | HME, HGA, and B2A · Qualitative counterfactual example | reason correctness | Hurt |
Closest papers by meaning
- Commonsense Video Question Answering through Video-Grounded Entailment Tree Reasoning (CVPR 2025 · similarity 0.91)
- VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection (CVPR 2025 · similarity 0.89)
- VITED: Video Temporal Evidence Distillation (CVPR 2025 · similarity 0.89)
- MoReVQA: Exploring Modular Reasoning Models for Video Question Answering (CVPR 2024 · similarity 0.88)
- Divide and Conquer: Exploring Language-centric Tree Reasoning for Video Question-Answering (ICML 2025 · similarity 0.88)
- Not Another Text Benchmark: Putting the “Visual" Back in Visual Question Answering for Large Video Models (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.