Saved in:
Bibliographic Details
Main Authors: Sheng, Yuan, Hao, Yanbin, Li, Chenxu, Wang, Shuo, He, Xiangnan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.20622
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912666375684096
author Sheng, Yuan
Hao, Yanbin
Li, Chenxu
Wang, Shuo
He, Xiangnan
author_facet Sheng, Yuan
Hao, Yanbin
Li, Chenxu
Wang, Shuo
He, Xiangnan
contents Long video understanding remains challenging due to its complex, diverse, and temporally scattered content. Although video large language models (Video-LLMs) can process videos lasting tens of minutes, applying them to truly long sequences is computationally prohibitive and often leads to unfocused or inconsistent reasoning. A promising solution is to select only the most informative frames, yet existing approaches typically ignore temporal dependencies or rely on unimodal evidence, limiting their ability to provide complete and query-relevant context. We propose a Semantic-Visual Consensus Evidence Selection (SeViCES) framework for effective and reliable long video understanding. SeViCES is training-free and model-agnostic, and introduces two key components. The Semantic-Visual Consensus Frame Selection (SVCFS) module selects frames through (1) a temporal-aware semantic branch that leverages LLM reasoning over captions, and (2) a cluster-guided visual branch that aligns embeddings with semantic scores via mutual information. The Answer Consensus Refinement (ACR) module further resolves inconsistencies between semantic- and visual-based predictions by fusing evidence and constraining the answer space. Extensive experiments on long video understanding benchmarks show that SeViCES consistently outperforms state-of-the-art methods in both accuracy and robustness, demonstrating the importance of consensus-driven evidence selection for Video-LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2510_20622
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SeViCES: Unifying Semantic-Visual Evidence Consensus for Long Video Understanding
Sheng, Yuan
Hao, Yanbin
Li, Chenxu
Wang, Shuo
He, Xiangnan
Computer Vision and Pattern Recognition
Long video understanding remains challenging due to its complex, diverse, and temporally scattered content. Although video large language models (Video-LLMs) can process videos lasting tens of minutes, applying them to truly long sequences is computationally prohibitive and often leads to unfocused or inconsistent reasoning. A promising solution is to select only the most informative frames, yet existing approaches typically ignore temporal dependencies or rely on unimodal evidence, limiting their ability to provide complete and query-relevant context. We propose a Semantic-Visual Consensus Evidence Selection (SeViCES) framework for effective and reliable long video understanding. SeViCES is training-free and model-agnostic, and introduces two key components. The Semantic-Visual Consensus Frame Selection (SVCFS) module selects frames through (1) a temporal-aware semantic branch that leverages LLM reasoning over captions, and (2) a cluster-guided visual branch that aligns embeddings with semantic scores via mutual information. The Answer Consensus Refinement (ACR) module further resolves inconsistencies between semantic- and visual-based predictions by fusing evidence and constraining the answer space. Extensive experiments on long video understanding benchmarks show that SeViCES consistently outperforms state-of-the-art methods in both accuracy and robustness, demonstrating the importance of consensus-driven evidence selection for Video-LLMs.
title SeViCES: Unifying Semantic-Visual Evidence Consensus for Long Video Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.20622