# ClusterChat: Multi-Feature Search for Corpus Exploration

arXiv: https://arxiv.org/abs/2412.14533 · HTML: https://mlsanity.com/paper/2412.14533/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 14 references that are in the map.
- First posted on arXiv 19 Dec 2024.

## Builds on (most cited first, 5 of 5)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing](https://mlsanity.com/paper/2007.15779/) (2020)
- [Embedding Projector: Interactive Visualization and Interpretation of Embeddings](https://mlsanity.com/paper/1611.05469/) (2016)
- [WizMap: Scalable Interactive Visualization for Exploring Large Machine Learning Embeddings](https://mlsanity.com/paper/2306.09328/) (2023)
- [Knowledge Navigator: LLM-guided Browsing Framework for Exploratory Search in Scientific Literature](https://mlsanity.com/paper/2408.15836/) (2024)

## Abstract (arXiv)

Exploring large-scale text corpora presents a significant challenge in biomedical, finance, and legal domains, where vast amounts of documents are continuously published. Traditional search methods, such as keyword-based search, often retrieve documents in isolation, limiting the user's ability to easily inspect corpus-wide trends and relationships. We present ClusterChat (The demo video and source code are available at: https://github.com/achouhan93/ClusterChat), an open-source system for corpus exploration that integrates cluster-based organization of documents using textual embeddings with lexical and semantic search, timeline-driven exploration, and corpus and document-level question answering (QA) as multi-feature search capabilities. We validate the system with two case studies on a four million abstract PubMed dataset, demonstrating that ClusterChat enhances corpus exploration by delivering context-aware insights while maintaining scalability and responsiveness on large-scale document collections.