Metraipolis › archive › dc_37869eab67f5
HeiCo-FOCUS: A Clinically Grounded Dataset for Long-Context Video Understanding — arXiv:2610.10156
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.10156
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T12:47:51.996Z
- SHA-256 of the archived text
7772dde18804bf9dfd699ca9960e2e30472bf14b68471883c533acd484db67b5- Stable URL of this page
https://metraipolis.com/documento/dc_37869eab67f5
How to cite
HeiCo-FOCUS: A Clinically Grounded Dataset for Long-Context Video Understanding — arXiv:2610.10156 — arXiv (en), https://arxiv.org/abs/2610.10156 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_37869eab67f5 (Metraipolis archive dc_37869eab67f5, sha256 7772dde18804bf9dfd699ca9960e2e30472bf14b68471883c533acd484db67b5)
BibTeX
@misc{dc37869eab67f5,
title = {HeiCo-FOCUS: A Clinically Grounded Dataset for Long-Context Video Understanding — arXiv:2610.10156},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.10156},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_37869eab67f5; sha256 7772dde18804bf9dfd699ca9960e2e30472bf14b68471883c533acd484db67b5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10156. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Recent advances in Vision-Language Models (VLMs) have led to rapid progress in video understanding across a wide range of benchmark tasks. However, existing evaluations largely focus on short-term reasoning, failing to assess a critical capability: maintaining cumulative temporal consistency over extended time horizons. To close this evaluation gap, we introduce HeiCo-FOCUS, a clinically grounded dataset for evaluating long-context video understanding through the task of Foreign Object Contextual Understanding in Surgery. Built on a dataset of Heidelberg Colorectal surgeries, this task requires models to continuously track multiple objects as they are inserted, manipulated, occluded, and removed over procedures lasting up to hours. HeiCo-FOCUS comprises 30,000 visual question answering (VQA) pairs covering five core capabilities: object recognition, temporal grounding, aggregation, event and procedural understanding, and complex reasoning. The dataset was constructed through a rigorous multi-stage annotation pipeline involving large-scale crowd annotation and 39 surgical domain experts to ensure high quality and clinical relevance. To systematically probe model behavior, we introduce a multi-track evaluation framework that progressively increases temporal and contextual demands from single frames to full procedures. Experiments with ten frontier VLMs show that HeiCo-FOCUS tasks are far from solved: only around half of the models clearly outperform a text-only baseline. Across the video tracks, models perform best on event and procedural understanding (mean Accuracy: 56.5% across all models), while temporal grounding remains particularly challenging for all evaluated models (mean Accuracy: 19.7%). We therefore expect HeiCo-FOCUS to serve as a catalyst for the development of models capable of reliable, temporally consistent reasoning over hours-long videos. Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2610.10156 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-08T12:18:15.453Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).