EgoJev: Finding Evidence in Egocentric Video by Reading VLM Decisions

Questions about long egocentric video often depend on one brief moment, such as the second a hand puts an object down. A vision-language model (VLM) that sees a few evenly spaced frames usually misses it. Training-free methods rank frames with a separate image-text model or ask the VLM to write a time. In a controlled test, the VLM itself locates the moment better when it chooses among labeled frames, and we read this choice from its output distribution. With the same candidates, this choice lands on the evidence more often than image-text scores or a written time span, and it compares all candidates in one decision. Based on this finding, we propose EgoJev, a training-free frame selector for wearable assistants. EgoJev reads where the evidence is from labeled thumbnails and looks again at the windows it points to. It places its few answer frames at the quantiles of the resulting distribution. One frozen VLM selects, answers and returns the evidence time, with no second model. A wearable device can encode its coarse thumbnails while it records. With Qwen3.5-4B and 8 answer frames, EgoJev raises QAEgo4D accuracy from 51.2% with uniform sampling to 62.0%, and it outperforms recent training-free frame selectors on QAEgo4D and MyEgo. A single-level version runs on a Mac and an iPhone.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-18
DOI
https://doi.org/10.48550/arxiv.2611.24616
Primary Topic
Multimodal Machine Learning Applications
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

EgoJev: Finding Evidence in Egocentric Video by Reading VLM Decisions

Hanspeter Pfister, Chunggi Lee
arXiv (Cornell University)
Multimodal Machine Learning Applications
preprint

EgoJev: Finding Evidence in Egocentric Video by Reading VLM Decisions

Hanspeter Pfister, Chunggi Lee
preprint en

Abstract

Questions about long egocentric video often depend on one brief moment, such as the second a hand puts an object down. A vision-language model (VLM) that sees a few evenly spaced frames usually misses it. Training-free methods rank frames with a separate image-text model or ask the VLM to write a time. In a controlled test, the VLM itself locates the moment better when it chooses among labeled frames, and we read this choice from its output distribution. With the same candidates, this choice lands on the evidence more often than image-text scores or a written time span, and it compares all candidates in one decision. Based on this finding, we propose EgoJev, a training-free frame selector for wearable assistants. EgoJev reads where the evidence is from labeled thumbnails and looks again at the windows it points to. It places its few answer frames at the quantiles of the resulting distribution. One frozen VLM selects, answers and returns the evidence time, with no second model. A wearable device can encode its coarse thumbnails while it records. With Qwen3.5-4B and 8 answer frames, EgoJev raises QAEgo4D accuracy from 51.2% with uniform sampling to 62.0%, and it outperforms recent training-free frame selectors on QAEgo4D and MyEgo. A single-level version runs on a Mac and an iPhone.

arXiv (Cornell University)
Multimodal Machine Learning Applications
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon