EgoAdapt: A Multi-Scene Egocentric Adaptation Method for CVPR 2026 HD-EPIC VQA Challenge

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Zhiwei, Hu, Yupeng, Li, Zixu, Fu, Zhiheng, Qiu, Guozhi, Guan, Weili, Nie, Liqiang
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911713159282688
author Chen, Zhiwei
Hu, Yupeng
Li, Zixu
Fu, Zhiheng
Qiu, Guozhi
Guan, Weili
Nie, Liqiang
author_facet Chen, Zhiwei
Hu, Yupeng
Li, Zixu
Fu, Zhiheng
Qiu, Guozhi
Guan, Weili
Nie, Liqiang
contents This technical report presents our solution, EgoAdapt (Egocentric Adaptation via Category, Calibration, and Consistency), to the CVPR 2026 HD-EPIC VQA challenge. HD-EPIC evaluates whether a vision-language model can reason over realistic first-person kitchen videos, where the evidence for an answer may be a short hand-object interaction, a long recipe trajectory, a spatial relation to a fixture, or a subtle gaze cue. The benchmark contains 26K multiple-choice questions across seven macro-categories: recipe, ingredient, nutrition, fine-grained action, 3D perception, object motion, and gaze. We observe that the main difficulty is not only model capacity, but also the mismatch between a single generic inference recipe and the heterogeneous temporal, spatial, and semantic structure of the benchmark. Our method, EgoAdapt, introduces three inference-time components: (1) category-conditioned routing with per-category prompts, frame budgets, and sampling rates; (2) calibrated option scoring that evaluates all candidate answers with letter-token likelihoods and generation agreement instead of relying only on direct generation; and (3) test-time consistency adaptation that aggregates predictions across option permutations and verification-style prompts for ambiguous cases. This design substantially improves over the available HD-EPIC baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2605_24500
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EgoAdapt: A Multi-Scene Egocentric Adaptation Method for CVPR 2026 HD-EPIC VQA Challenge
Chen, Zhiwei
Hu, Yupeng
Li, Zixu
Fu, Zhiheng
Qiu, Guozhi
Guan, Weili
Nie, Liqiang
Computer Vision and Pattern Recognition
This technical report presents our solution, EgoAdapt (Egocentric Adaptation via Category, Calibration, and Consistency), to the CVPR 2026 HD-EPIC VQA challenge. HD-EPIC evaluates whether a vision-language model can reason over realistic first-person kitchen videos, where the evidence for an answer may be a short hand-object interaction, a long recipe trajectory, a spatial relation to a fixture, or a subtle gaze cue. The benchmark contains 26K multiple-choice questions across seven macro-categories: recipe, ingredient, nutrition, fine-grained action, 3D perception, object motion, and gaze. We observe that the main difficulty is not only model capacity, but also the mismatch between a single generic inference recipe and the heterogeneous temporal, spatial, and semantic structure of the benchmark. Our method, EgoAdapt, introduces three inference-time components: (1) category-conditioned routing with per-category prompts, frame budgets, and sampling rates; (2) calibrated option scoring that evaluates all candidate answers with letter-token likelihoods and generation agreement instead of relying only on direct generation; and (3) test-time consistency adaptation that aggregates predictions across option permutations and verification-style prompts for ambiguous cases. This design substantially improves over the available HD-EPIC baselines.
title EgoAdapt: A Multi-Scene Egocentric Adaptation Method for CVPR 2026 HD-EPIC VQA Challenge
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.24500