<link rel="stylesheet" href="styles.cf390c6bfbe7555f.css">
Repository logo

MoM: Multi-to-Multi Identification of Decoupled Multifaceted Clustering with Dynamic Representation Fusion

aut.relation.conferenceMM '26: The 34th ACM International Conference on Multimedia
aut.relation.endpage14625
aut.relation.startpage14616
dc.contributor.authorZhang, Peipei
dc.contributor.authorZhu, Wenjie
dc.contributor.authorZhou, Qinli
dc.contributor.authorYan, Wei Qi
dc.date.accessioned2026-11-19T22:03:02Z
dc.date.issued2026-11-09
dc.description.abstractMultifaceted clustering is dedicated to excavating diverse and heterogeneous facets from complex multimedia data, which is a critical task for multimedia content understanding. Existing approaches primarily focus on decomposing a single dataset into multiple subspaces or sequentially generating alternative clusterings. However, these methods are severely constrained by the inherent limited information contained in the data itself, resulting in suboptimal clustering performance and indistinct facet mining results. To address the issue, we propose a novel paradigm shift from Single-to-Multi to Multi-to-Multi (MoM) clustering for multimedia data, which proactively integrates multiple informative views for decoupled multifaceted discovery, breaking the bottleneck of single-source data information scarcity. The proposed MoM framework consists of two core modules: the Dynamic Representation Fusion (DRF) module and the Decoupled Multifaceted Clustering (DMC) module. Specifically, the DRF module implements an adaptive fusion strategy that generates and dynamically fuses diverse representations guided by the anchors derived from a pretrained model, ensuring both fidelity and diversity. The DMC module leverages a structured self-expression model to jointly learn shared and facet-specific affinity matrices, producing well-separated and interpretable clusters. Unlike existing methods that passively decompose limited data into disjoint subspaces, our end-to-end framework actively employs a clustering-oriented representation fusion strategy to learn decoupled features tailored to each clustering facet. Extensive experiments show that MoM consistently outperforms state-of-the-art multifaceted clustering methods, validating the effectiveness of our MoM framework.
dc.identifier.citationMM '26: Proceedings of the 34th ACM International Conference on Multimedia. Rio de Janeiro, Brazil. November 10 - 14, 2026. Pages 14616 - 14625. ISBN: 9798400722134 DOI: 10.1145/3767308
dc.identifier.doi10.1145/3767308.3835249
dc.identifier.urihttp://hdl.handle.net/10292/22442
dc.publisherACM
dc.relation.urihttps://dl.acm.org/doi/10.1145/3767308.3835249
dc.rights© 2026 Owner/Author. Open access.
dc.rights.accessrightsOpenAccess
dc.rights.licenseCreative Commons Attribution 4.0
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subject46 Information and Computing Sciences
dc.subject4611 Machine Learning
dc.subjectNetworking and Information Technology R&D (NITRD)
dc.subjectdeep multifaceted clustering
dc.subjectdynamic representation fusion
dc.subjectdecoupled representation learning
dc.titleMoM: Multi-to-Multi Identification of Decoupled Multifaceted Clustering with Dynamic Representation Fusion
dc.typeConference Contribution
pubs.elements-id778356

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
ACMMM26.pdf
Size:
11.66 MB
Format:
Adobe Portable Document Format
Description:
Conference paper