SOURCE-LINKED INTELLIGENCE

Beyond Top Words: MonoTM for Topic Modeling with Interpretable Monosemantic Features

arXiv · AI, language, vision and robotics · article · Sep 9, 2026 · UTC

Topic models summarize large text corpora, but top-ranked words often provide only a limited representation of topic semantics. Sparse autoencoders (SAEs) offer a way to move beyond word-level descriptors by extracting interpretable features from dense representations, yet how feature interpretability relates to topic-inference quality remains unclear. We introduce \textbf{MonoTM}, an interpretable topic modeling framework that decouples these roles. Across three benchmark corpora, we show that document--topic mixture estimation and semantic interpretation favor different SAE configurations an

Read original source ↗ Open in workspace

recordType

paper

region

Global