# Latent Tree Models for Hierarchical Topic Detection

arXiv: https://arxiv.org/abs/1605.06650 · HTML: https://mlsanity.com/paper/1605.06650/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 May 2016.

## Built on by (most cited first, 6 of 6)

- [Emergence of Hierarchical Emotion Organization in Large Language Models](https://mlsanity.com/paper/2507.10599/) (2025)
- [Two-Stage Holistic and Contrastive Explanation of Image Classification](https://mlsanity.com/paper/2306.06339/) (2023)
- [Modality Matching Matters: Calibrating Language Distances for Cross-Lingual Transfer in URIEL+](https://mlsanity.com/paper/2510.19217/) (2025)
- [Analysis and tuning of hierarchical topic models based on Renyi entropy approach](https://mlsanity.com/paper/2101.07598/) (2021)
- [AceMap: Knowledge Discovery through Academic Graph](https://mlsanity.com/paper/2403.02576/) (2024)
- [CASE: Contrastive Activation for Class-Sensitive Explanations](https://mlsanity.com/paper/2506.07327/) (2025)

## Abstract (arXiv)

We present a novel method for hierarchical topic detection where topics are obtained by clustering documents in multiple ways. Specifically, we model document collections using a class of graphical models called hierarchical latent tree models (HLTMs). The variables at the bottom level of an HLTM are observed binary variables that represent the presence/absence of words in a document. The variables at other levels are binary latent variables, with those at the lowest latent level representing word co-occurrence patterns and those at higher levels representing co-occurrence of patterns at the level below. Each latent variable gives a soft partition of the documents, and document clusters in the partitions are interpreted as topics. Latent variables at high levels of the hierarchy capture long-range word co-occurrence patterns and hence give thematically more general topics, while those at low levels of the hierarchy capture short-range word co-occurrence patterns and give thematically more specific topics. Unlike LDA-based topic models, HLTMs do not refer to a document generation process and use word variables instead of token variables. They use a tree structure to model the relationships between topics and words, which is conducive to the discovery of meaningful topics and topic hierarchies.