# Disentangling Similarity and Relatedness in Topic Models

Source: [Disentangling Similarity and Relatedness in Topic Models](<https://www.sota2.com/research/paper/69b248afd030084c22e2fa73>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-08-04T05:00:05.533Z

Authors: Hanlin Xiao, Yang Wang, Mauricio A. \\'Alvarez, Rainer Breitling

Published: 2026-03-11T10:27:43.000Z

## Abstract

The recent success of large pre-trained language models (PLMs) has motivated their integration into topic modeling. However, PLM-augmented topic models differ from classical co-occurrence models such as Latent Dirichlet Allocation (LDA) not only in performance, but also in the type of semantic structure they capture. We formalize this distinction along two psycholinguistic axes: thematic relatedness (dog/bone) and taxonomic similarity (dog/wolf). To measure both axes over topic words, we construct a large synthetic benchmark of word pairs using LLM-based annotation and train a neural scorer on it. Across multiple corpora and model families, the scorer places different topic-model families at distinct positions within the joint similarity-relatedness space. The two scores further predict downstream task performance: tasks requiring similarity benefit from similarity-rich topics, whereas tasks requiring relatedness benefit from the converse, and excessive emphasis on either axis degrades performance on tasks aligned with the opposing semantic structure. Neither axis is uniformly beneficial. Measuring both therefore provides a practical, model-agnostic diagnostic for evaluating the semantic structure captured by topic models.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Word Relatedness Prediction on TxThmNorms 651 pairs (test)](<https://www.sota2.com/research/sota/word-relatedness-prediction-on-txthmnorms-651-pairs-test.md>): Word Relatedness Prediction; TxThmNorms 651 pairs (test); Spearman Correlation: 0.343.

- [Word Similarity Prediction on TxThmNorms 651 pairs (test)](<https://www.sota2.com/research/sota/word-similarity-prediction-on-txthmnorms-651-pairs-test.md>): Word Similarity Prediction; TxThmNorms 651 pairs (test); Spearman Correlation: 0.536.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.10619>)

[PDF](<https://arxiv.org/pdf/2603.10619.pdf>)