Hierarchical Graph Topic Modeling with Topic Tree-based Transformer

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Delvin Ce, Yang, Menglin, Wu, Xiaobao, Zhang, Jiasheng, Lauw, Hady W.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916617706799104
author Zhang, Delvin Ce
Yang, Menglin
Wu, Xiaobao
Zhang, Jiasheng
Lauw, Hady W.
author_facet Zhang, Delvin Ce
Yang, Menglin
Wu, Xiaobao
Zhang, Jiasheng
Lauw, Hady W.
contents Textual documents are commonly connected in a hierarchical graph structure where a central document links to others with an exponentially growing connectivity. Though Hyperbolic Graph Neural Networks (HGNNs) excel at capturing such graph hierarchy, they cannot model the rich textual semantics within documents. Moreover, text contents in documents usually discuss topics of different specificity. Hierarchical Topic Models (HTMs) discover such latent topic hierarchy within text corpora. However, most of them focus on the textual content within documents, and ignore the graph adjacency across interlinked documents. We thus propose a Hierarchical Graph Topic Modeling Transformer to integrate both topic hierarchy within documents and graph hierarchy across documents into a unified Transformer. Specifically, to incorporate topic hierarchy within documents, we design a topic tree and infer a hierarchical tree embedding for hierarchical topic modeling. To preserve both topic and graph hierarchies, we design our model in hyperbolic space and propose Hyperbolic Doubly Recurrent Neural Network, which models ancestral and fraternal tree structure. Both hierarchies are inserted into each Transformer layer to learn unified representations. Both supervised and unsupervised experiments verify the effectiveness of our model.
format Preprint
id arxiv_https___arxiv_org_abs_2502_11345
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hierarchical Graph Topic Modeling with Topic Tree-based Transformer
Zhang, Delvin Ce
Yang, Menglin
Wu, Xiaobao
Zhang, Jiasheng
Lauw, Hady W.
Computation and Language
Textual documents are commonly connected in a hierarchical graph structure where a central document links to others with an exponentially growing connectivity. Though Hyperbolic Graph Neural Networks (HGNNs) excel at capturing such graph hierarchy, they cannot model the rich textual semantics within documents. Moreover, text contents in documents usually discuss topics of different specificity. Hierarchical Topic Models (HTMs) discover such latent topic hierarchy within text corpora. However, most of them focus on the textual content within documents, and ignore the graph adjacency across interlinked documents. We thus propose a Hierarchical Graph Topic Modeling Transformer to integrate both topic hierarchy within documents and graph hierarchy across documents into a unified Transformer. Specifically, to incorporate topic hierarchy within documents, we design a topic tree and infer a hierarchical tree embedding for hierarchical topic modeling. To preserve both topic and graph hierarchies, we design our model in hyperbolic space and propose Hyperbolic Doubly Recurrent Neural Network, which models ancestral and fraternal tree structure. Both hierarchies are inserted into each Transformer layer to learn unified representations. Both supervised and unsupervised experiments verify the effectiveness of our model.
title Hierarchical Graph Topic Modeling with Topic Tree-based Transformer
topic Computation and Language
url https://arxiv.org/abs/2502.11345