Research map / Video / Video-language understanding
EgoTaskQA: Understanding Human Tasks in Egocentric Videos
Read the paperCheck 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 |
|---|---|---|---|
| remove vision branch (language-only) (ablation) | HCRN · EgoTaskQA normal, all | accuracy: 42.20 -> 32.51 | Hurt |
| remove vision (BERT vs VisualBERT) (ablation) | VisualBERT · EgoTaskQA normal, all | accuracy: 37.93 -> 33.92 | Hurt |
| remove vision (ablation) | HCRN · EgoTaskQA normal, object questions | accuracy: 45.31 -> 6.33 | Hurt |
| remove vision (ablation) | HCRN · EgoTaskQA normal, change questions | accuracy: 67.38 -> 68.77 | No clear effect |
| remove object-information obfuscation in action text (ablation) | video QA models · EgoTaskQA normal | accuracy | Helped |
| switch from normal to indirect split | HCRN · EgoTaskQA all | accuracy: 42.20 -> 41.56 | Hurt |
| switch from normal to indirect split | ClipBERT · EgoTaskQA all | accuracy: 39.87 -> 24.08 | Hurt |
| switch from normal to indirect split | PSAC · EgoTaskQA all | accuracy: 38.90 -> 42.25 | Helped |
| switch from normal to indirect split | BERT language-only · EgoTaskQA change questions | accuracy | Hurt |
Closest papers by meaning
- EgoExoBench: A Benchmark for First- and Third-person View Video Understanding in MLLMs (NeurIPS 2025 · similarity 0.91)
- EgoThinker: Unveiling Egocentric Reasoning with Spatio-Temporal CoT (NeurIPS 2025 · similarity 0.91)
- A Backpack Full of Skills: Egocentric Video Understanding with Diverse Task Perspectives (CVPR 2024 · similarity 0.90)
- Omnia de EgoTempo: Benchmarking Temporal Understanding of Multi-Modal LLMs in Egocentric Videos (CVPR 2025 · similarity 0.89)
- Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs (NeurIPS 2026 · similarity 0.89)
- In the Eye of MLLM: Benchmarking Egocentric Video Intent Understanding with Gaze-Guided Prompting (NeurIPS 2025 · similarity 0.89)
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.