Papers
arxiv:2605.27800

CuriosAI Submission to the CASTLE Challenge at EgoVis 2026

Published on May 27
Authors:
,
,

Abstract

Two approaches are evaluated for answering questions from multi-view egocentric video: SVA uses a search-verify-answer pipeline with VLM verification and LLM judgment, while TMKG constructs a temporal multimodal knowledge graph for question answering.

CASTLE 2026 asks 185 multiple-choice questions over 600+ hours of synchronized multi-view egocentric video. We explore two approaches on top of a shared multimodal preprocessing layer, including per-person timelines, speaker-resolved transcripts, and multi-VLM caption ensembles. Approach A, SVA: Search-Verify-Answer, is a three-stage pipeline that hierarchically narrows to a primary window, verifies sub-windows with a VLM under four anti-confabulation rules, and fuses evidence with an LLM judge under an evidence-priority hierarchy. Approach B, TMKG: Temporal-Multimodal-Knowledge-Graph, is the contrast: it builds a temporal multimodal knowledge graph, locates a primary cell via graph search, and produces the final answer with a single grounded VLM. SVA reaches a leaderboard accuracy of 0.50 and is our final challenge submission; TMKG reaches 0.35.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2605.27800
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2605.27800 in a model README.md to link it from this page.

Datasets citing this paper 1

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2605.27800 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.