AI Paper Atlas

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

ICCV 2025 · Video › Video-language understanding · approach: Dense video captioning · 2 citations (OpenAlex)

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 changedWhereMeasuredEffect
add window-based video-narration alignment loss (ablation)HiERO (EgoVLP) · EgoProceL averageF1 (%): 40.0 -> 43.8Improved
add alignment loss (ablation)HiERO (EgoVLP) · EgoMCQ intraaccuracy (%): 57.2 -> 59.5Improved
add alignment loss (ablation)HiERO (EgoVLP) · Goal-Step step grounding (zero-shot)R@1 (%): 10.73 -> 11.27Improved
add training-time functional-thread clustering (ablation)HiERO (EgoVLP) · EgoProceL averageF1 (%): 43.8 -> 43.3Hurt
add training-time functional-thread clustering (ablation)HiERO (EgoVLP) · Goal-Step step grounding (zero-shot)R@5 (%): 27.35 -> 27.12Hurt
add functional threads loss (ablation)HiERO (EgoVLP) · EgoProceL averageF1 (%): 43.3 -> 44.5Improved
add functional threads loss (ablation)HiERO (EgoVLP) · Goal-Step step grounding (zero-shot)R@1 (%): 11.44 -> 11.57Improved
use HiERO features instead of Omnivore for zero-shot clusteringHiERO (Omnivore) · EgoProceL averageF1 (%): 39.1 -> 44.0Improved
use HiERO features instead of EgoVLP for zero-shot clusteringHiERO (EgoVLP) · EgoProceL averageF1 (%): 40.0 -> 44.5Improved
use HiERO features as input to VSLNetHiERO (EgoVLP) · Goal-Step step grounding, supervisedR@1 mIoU@0.3 (%): 15.43 -> 15.64Improved
use HiERO features instead of EgoVLP for zero-shot step localizationHiERO (EgoVLP) · Goal-Step step localizationmAP avg: 8.3 -> 8.7Improved
use spectral instead of K-means clustering (ablation)HiERO (EgoVLP) · Goal-Step step grounding, zero-shotR@1 mIoU@0.3 (%): 10.35 -> 11.57Improved

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy