Search papers, labs, and topics across Lattice.
CobwebTM is introduced as a novel lifelong hierarchical topic model that leverages incremental probabilistic concept formation. It adapts the Cobweb algorithm to continuous document embeddings, enabling online construction of semantic hierarchies. Experiments across diverse datasets demonstrate CobwebTM's strong topic coherence, temporal stability, and high-quality hierarchical organization, showing the effectiveness of combining incremental symbolic concept formation with pretrained representations for topic modeling.
Forget retraining: CobwebTM builds interpretable topic hierarchies on the fly, adapting to streaming data without catastrophic forgetting.
Topic modeling seeks to uncover latent semantic structure in text corpora with minimal supervision. Neural approaches achieve strong performance but require extensive tuning and struggle with lifelong learning due to catastrophic forgetting and fixed capacity, while classical probabilistic models lack flexibility and adaptability to streaming data. We introduce \textsc{CobwebTM}, a low-parameter lifelong hierarchical topic model based on incremental probabilistic concept formation. By adapting the Cobweb algorithm to continuous document embeddings, \textsc{CobwebTM} constructs semantic hierarchies online, enabling unsupervised topic discovery, dynamic topic creation, and hierarchical organization without predefining the number of topics. Across diverse datasets, \textsc{CobwebTM} achieves strong topic coherence, stable topics over time, and high-quality hierarchies, demonstrating that incremental symbolic concept formation combined with pretrained representations is an efficient approach to topic modeling.
