BERTopic · Issues· 474 open

About Coherence of topic models

Topic deletion fails if topics have custom lables

BadRequestError: Error code: 400 - {'error': {'message': "Unsupported parameter: 'stop' is not supported with this model.", 'type': 'invalid_request_error', 'param': 'stop', 'code': 'unsupported_parameter'}}

Topic aspects are not updated when merging topics and the model uses a custom vectorizer

Creating topics over time fails if model has deleted topics

Creating Topic Taxonomies

Add structured multi aspect output to the LangChain representation 2:06 AM

Better Topic Modeling Visualization/Saving for Image Only Analysis

`MaximalMarginalRelevance` makes 2N embedding calls instead of 1

Add top-down (divisive) strategy for `hierarchical_topics()`

Support pre-tokenized documents to avoid redundant tokenization across pipeline stages

Add contrastive and embedding-aware feature importance as representation aspects

Lightweight soft topic assignment via temperature-scaled embedding distances

Support reusing pre-computed UMAP embeddings across runs (grid search, caching)

`hierarchical_topics()` parent nodes show raw c-TF-IDF keywords instead of representation model labels