NeuroMax: Enhancing Neural Topic Modeling via Maximizing Mutual Information and Group Topic Regularization

TL;DR

NeuroMax introduces a neural topic modeling framework that maximizes mutual information with pre-trained language models and employs group regularization to improve topic coherence, reduce inference time, and enhance downstream task performance.

Contribution

It proposes a novel NeuroMax framework combining mutual information maximization and group topic regularization, addressing inference efficiency and topic relationship modeling in neural topic models.

Findings

Reduces inference time significantly.

Produces more coherent and interpretable topics.

Enhances downstream task performance with better document embeddings.

Abstract

Recent advances in neural topic models have concentrated on two primary directions: the integration of the inference network (encoder) with a pre-trained language model (PLM) and the modeling of the relationship between words and topics in the generative model (decoder). However, the use of large PLMs significantly increases inference costs, making them less practical for situations requiring low inference times. Furthermore, it is crucial to simultaneously model the relationships between topics and words as well as the interrelationships among topics themselves. In this work, we propose a novel framework called NeuroMax (Neural Topic Model with Maximizing Mutual Information with Pretrained Language Model and Group Topic Regularization) to address these challenges. NeuroMax maximizes the mutual information between the topic representation obtained from the encoder in neural topic models and the representation derived from the PLM. Additionally, NeuroMax employs optimal transport to learn the relationships between topics by analyzing how information is transported among them. Experimental results indicate that NeuroMax reduces inference time, generates more coherent topics and topic groups, and produces more representative document embeddings, thereby enhancing performance on downstream tasks.