Pre-trained language models have led to a new state-of-the-art in many NLP tasks. However, for topic modeling, statistical generative models such as LDA are still prevalent, which do not easily allow incorporating contextual word vectors. They might yield topics that do not align very well with human judgment. In this work, we propose a novel topic modeling and inference algorithm. We suggest a bag of sentences (BoS) approach using sentences as the unit of analysis. We leverage pre-trained sentence embeddings by combining generative process models with clustering. We derive a fast inference algorithm based on expectation maximization, hard assignments, and an annealing process. Our evaluation shows that our method yields state-of-the art results with relatively little computational demands. Our methods is more flexible compared to prior works leveraging word embeddings, since it provides the possibility to customize topic-document distributions using priors. Code is at \url{https://github.com/JohnTailor/BertSenClu}.

A few days ago I found out that there had appeared lda2vec (by Chris Moody) – a hybrid algorithm combining best ideas from well-known LDA (Latent Dirichlet Allocation) topic modeling algorithm and from a bit less well-known tool for language modeling named word2vec. And now I'm going to tell you a tale about lda2vec and my attempts to try it and compare with simple LDA implementation (I used gensim package for this). What is cool about it? It means that LDA is able to create document (and topic) representations that are not so flexible but mostly interpretable to humans. Also, LDA treats a set of documents as a set of documents, whereas word2vec works with a set of documents as with a very long text string.