Research map / Video / Video-language understanding
FunQA: Towards Surprising Video Comprehension
Binzhu Xie, Sicheng Zhang, Zitang Zhou, Bo Li, Yuanhan Zhang, Jack Hessel, Jingkang Yang, Ziwei Liu
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 FunMentor multi-turn refinement agent to Video-ChatGPT | Video-ChatGPT · HumorQA H2 description | GPT-4 score: 24.3 -> 33.2 | Improved |
| add FunMentor to Video-ChatGPT | Video-ChatGPT · HumorQA H4 title | GPT-4 score: 41.2 -> 65.1 | Improved |
| add FunMentor to Video-ChatGPT | Video-ChatGPT · HumorQA H3 reasoning | GPT-4 score: 24.9 -> 36.5 | Improved |
| add FunMentor to Otter (FunQA) | Otter (FunQA) · HumorQA H3 reasoning | GPT-4 score: 31.7 -> 45.8 | Improved |
| add FunMentor to Otter (FunQA) | Otter (FunQA) · CreativeQA C5 scoring | accuracy: 69.4 -> 69.4 | No clear effect |
| add FunMentor to Otter (FunQA) | Otter (FunQA) · HumorQA H2 description | BLEURT: 38.4 -> 33.4 | Hurt |
| fine-tune Otter on FunQA instead of Dense Caption | Otter · FunQA free-text (10% sample) | GPT-4 score: 22 -> 28 | Improved |
| use timestamp-capable TimeChat versus mean time span baseline | localization · HumorQA H1 | IOU: 55.5 -> 8.4 | Hurt |
| evaluate with WUPS versus GPT-4 metric | Otter (FunQA) · NExT-OE | WUPS vs GPT-4: 0.79 WUPS vs 73.06 GPT-4 | Mixed |
Closest papers by meaning
- VRR-QA: Visual Relational Reasoning in Videos Beyond Explicit Cues (CVPR 2026 · similarity 0.91)
- ANetQA: A Large-Scale Benchmark for Fine-Grained Compositional Reasoning Over Untrimmed Videos (CVPR 2023 · similarity 0.90)
- CG-Bench: Clue-grounded Question Answering Benchmark for Long Video Understanding (ICLR 2025 · similarity 0.89)
- VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection (CVPR 2025 · similarity 0.89)
- Commonsense Video Question Answering through Video-Grounded Entailment Tree Reasoning (CVPR 2025 · similarity 0.89)
- MMBench-Video: A Long-Form Multi-Shot Benchmark for Holistic Video Understanding (NeurIPS 2024 · 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.
