Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

6-2026

Abstract

There are various factors affecting the performance of video search. An imprecise query will enlarge search space and reduce the discriminative power of ranking functions. This problem is further exacerbated by the presence of numerous visually or semantically similar videos in large datasets. Consequently, users need to painstakingly browse through many highly similar candidates to locate the search target, leading to increased cognitive load and inefficient searching. Ideally, engaging users through interactive questioning to resolve uncertainties in the search process is an effective strategy for progressively narrowing down the search space. However, despite rapid advances in deep learning, generating informative questions conditioned on the user query and search result remains a highly difficult problem.In this paper, we propose a simple yet effective method to select informative concepts based on Shannon’s information theory for question composition. Instead of relying on large language models to generate questions, which is computationally slow and is subject to hallucinations, our approach selects the most discriminative concepts as questions to quickly prune the search space. We further provide practical insights into the implementation of this approach, specifically analyzing search result updates and top-K rank list sampling. Our analysis accounts for realistic constraints where the search engine may imperfectly index video content and users may provide inaccurate or misleading answers. Despite its conceptual simplicity, our method demonstrates strong retrieval performance on the Audio Visual Scene-Aware Dialog (AVSD) and TRECVid benchmarks.

Keywords

Question-answering system, Interactive video retrieval, Conversational search

Discipline

Databases and Information Systems | Graphics and Human Computer Interfaces

Research Areas

Intelligent Systems and Optimization

Areas of Excellence

Digital transformation

Publication

ICMR '26: Proceedings of the 2026 International Conference on Multimedia Retrieval, Amsterdam, The Netherlands, June 16-19

First Page

1759

Last Page

1767

ISBN

9798400726170

Identifier

10.1145/3805622.3810720

Publisher

ACM

City or Country

New York

Additional URL

https://doi.org/10.1145/3805622.3810720

Download

Share

COinS