FASTopic: Pretrained Transformer is a Fast, Adaptive, Stable, and Transferable Topic Model

TL;DR

FASTopic introduces a novel, efficient, and stable topic modeling approach using a pretrained Transformer with dual semantic-relation reconstruction and optimal transport regularization, outperforming existing methods.

Contribution

It proposes FASTopic, a new paradigm combining DSR and ETP for improved effectiveness, efficiency, stability, and transferability in topic modeling.

Findings

Outperforms state-of-the-art baselines on benchmark datasets.

Demonstrates superior efficiency and stability in various scenarios.

Shows enhanced transferability and adaptivity in topic modeling.

Abstract

Topic models have been evolving rapidly over the years, from conventional to recent neural models. However, existing topic models generally struggle with either effectiveness, efficiency, or stability, highly impeding their practical applications. In this paper, we propose FASTopic, a fast, adaptive, stable, and transferable topic model. FASTopic follows a new paradigm: Dual Semantic-relation Reconstruction (DSR). Instead of previous conventional, VAE-based, or clustering-based methods, DSR directly models the semantic relations among document embeddings from a pretrained Transformer and learnable topic and word embeddings. By reconstructing through these semantic relations, DSR discovers latent topics. This brings about a neat and efficient topic modeling framework. We further propose a novel Embedding Transport Plan (ETP) method. Rather than early straightforward approaches, ETP explicitly regularizes the semantic relations as optimal transport plans. This addresses the relation bias issue and thus leads to effective topic modeling. Extensive experiments on benchmark datasets demonstrate that our FASTopic shows superior effectiveness, efficiency, adaptivity, stability, and transferability, compared to state-of-the-art baselines across various scenarios.