Beyond Looking Up, Try Looking Around: Harmonizing Global Structure and Local Consistency in Optimal Transport for Short Text Clustering
2026-07-12 • Machine Learning
Machine Learning
AI summaryⓘ
The authors address the problem of grouping short text snippets into meaningful clusters. They point out that existing methods sometimes give different labels to texts that mean similar things, which hurts the clustering results. To fix this, they create a method that pays special attention to how similar samples are connected. Their approach uses this information to make better guesses (pseudo-labels) about which texts belong together, improving the overall clustering. They tested their method and found it works better than earlier techniques.
short text clusteringpseudo-labelingoptimal transportsemantic consistencyattention mechanismneighborhood semantic awarenessunsupervised learningcluster structure
Authors
Zhihao Yao, Yuxuan Gu, Jixuan Yin, Bo Li
Abstract
Pseudo-labeling based on Optimal Transport (OT) has become an effective mechanism for enhancing short text clustering. Existing OT methods are short in modeling semantic consistencies between samples, which may assign different pseudo-labels to semantically similar samples. These erroneous pseudo-labels can cause the model to produce inferior clusters. This paper proposes a novel short text clustering framework, which remedies the neglect of semantic consistency in existing OT methods, generating reliable pseudo-labels to facilitate clustering. Specifically, the proposed approach first designs an instance-level attention mechanism to capture semantic relationships between samples, which are then integrated into the OT formulation to endow the transport process with neighborhood semantic awareness. By solving the proposed OT formulation, reliable pseudo-labels are obtained that simultaneously account for sample-to-sample semantic consistency and sample-to-cluster global structure information. These pseudo-labels are then used as supervisory signals to guide the model to achieve accurate clustering. Extensive experiments demonstrate that the proposed approach outperforms state-of-the-art methods. The code is available at: \href{https://github.com/YZH0905/CAOT-STC}{https://github.com/YZH0905/CAOT-STC}.