Label Semantic Expansion via Label Guided Neural Topic Modeling
2026-08-31 • Computation and Language
Computation and LanguageArtificial Intelligence
AI summaryⓘ
The authors study how to improve analyzing collections of documents using labels instead of general hidden topics. They propose a new method called Label Semantic Expansion (LSE) that makes label descriptions better by adding topic words related to the documents. To do this well, they create a Label-Guided Neural Topic Model (LGNTM) that aligns topics with labels and connects them to word meanings and document content. Their experiments show that this approach works well for matching labels with topics, expanding label meanings, generating good topics, and helping with classification tasks.
topic modelslabel-aware modelingLabel Semantic Expansion (LSE)Label-Guided Neural Topic Model (LGNTM)topic-label alignmentsemantic representationdocument classificationlatent topicscorpus analysisneural topic modeling
Authors
Haojia Zheng, Yuyin Lu, Juntian Huang, Fan Ou, Yanghui Rao, Haoran Xie, Fu Lee Wang
Abstract
Topic models are widely used for content analysis, where users often analyze corpora around predefined labels rather than unordered latent topics. Existing label-aware topic models mainly follow a labels-for-topics perspective, using labels to guide topic learning, while the learned topics are not directly usable for label-centered analysis. We explore the reverse topics-for-labels perspective and instantiate it as Label Semantic Expansion (LSE), which enriches sparse label representations with corpus-grounded descriptive topic words. To exploit topics in LSE effectively, we propose a Label-Guided Neural Topic Model (LGNTM), which learns dedicated label-aligned topics, grounds them in lexical and document semantic spaces, and preserves consistency between topic structures and label structures. Experiments on label-topic alignment, label expansion, topic quality, and downstream classification demonstrate strong overall performance across complementary evaluation dimensions.