Research map / Video / Video-language understanding
HiERO: Understanding the Hierarchy of Human Behavior Enhances Reasoning on Egocentric Videos
Simone Alberto Peirone, Francesca Pistilli, Giuseppe Averta
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 window-based video-narration alignment loss (ablation) | HiERO (EgoVLP) · EgoProceL average | F1 (%): 40.0 -> 43.8 | Improved |
| add alignment loss (ablation) | HiERO (EgoVLP) · EgoMCQ intra | accuracy (%): 57.2 -> 59.5 | Improved |
| add alignment loss (ablation) | HiERO (EgoVLP) · Goal-Step step grounding (zero-shot) | R@1 (%): 10.73 -> 11.27 | Improved |
| add training-time functional-thread clustering (ablation) | HiERO (EgoVLP) · EgoProceL average | F1 (%): 43.8 -> 43.3 | Hurt |
| add training-time functional-thread clustering (ablation) | HiERO (EgoVLP) · Goal-Step step grounding (zero-shot) | R@5 (%): 27.35 -> 27.12 | Hurt |
| add functional threads loss (ablation) | HiERO (EgoVLP) · EgoProceL average | F1 (%): 43.3 -> 44.5 | Improved |
| add functional threads loss (ablation) | HiERO (EgoVLP) · Goal-Step step grounding (zero-shot) | R@1 (%): 11.44 -> 11.57 | Improved |
| use HiERO features instead of Omnivore for zero-shot clustering | HiERO (Omnivore) · EgoProceL average | F1 (%): 39.1 -> 44.0 | Improved |
| use HiERO features instead of EgoVLP for zero-shot clustering | HiERO (EgoVLP) · EgoProceL average | F1 (%): 40.0 -> 44.5 | Improved |
| use HiERO features as input to VSLNet | HiERO (EgoVLP) · Goal-Step step grounding, supervised | R@1 mIoU@0.3 (%): 15.43 -> 15.64 | Improved |
| use HiERO features instead of EgoVLP for zero-shot step localization | HiERO (EgoVLP) · Goal-Step step localization | mAP avg: 8.3 -> 8.7 | Improved |
| use spectral instead of K-means clustering (ablation) | HiERO (EgoVLP) · Goal-Step step grounding, zero-shot | R@1 mIoU@0.3 (%): 10.35 -> 11.57 | 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.
- EgoVLPv2 — EgoVLPv2: Egocentric Video-Language Pre-training with Fusion in the Backbone (ICCV 2023) · better on 4 of 4 shared results
- OPEL — OPEL: Optimal Transport Guided ProcedurE Learning (NeurIPS 2024) · better on 2 of 2 shared results
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
- Learning to Evolve Scenes: Reasoning about Human Activities with Scene Graphs (NeurIPS 2026 · similarity 0.87)
- Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention (NeurIPS 2025 · similarity 0.87)
- EgoThinker: Unveiling Egocentric Reasoning with Spatio-Temporal CoT (NeurIPS 2025 · similarity 0.87)
- Omnia de EgoTempo: Benchmarking Temporal Understanding of Multi-Modal LLMs in Egocentric Videos (CVPR 2025 · similarity 0.87)
- EgoExoBench: A Benchmark for First- and Third-person View Video Understanding in MLLMs (NeurIPS 2025 · similarity 0.87)
- Action Scene Graphs for Long-Form Understanding of Egocentric Videos (CVPR 2024 · 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.
