TWIST: Training-free and Label-free Short Text Clustering through Iterative Vector Updating with LLMs
This addresses the need for scalable, low-resource clustering in commercial settings like customer-facing chatbots, where labeled data is unavailable, though it is incremental as it builds on existing embedders.
The paper tackles the problem of clustering short texts without labeled data or prior knowledge of cluster counts, using iterative vector updating with LLMs, achieving comparable or superior results to state-of-the-art methods that rely on contrastive learning.
In this paper, we propose a training-free and label-free method for short text clustering that can be used on top of any existing embedder. In the context of customer-facing chatbots, companies are dealing with large amounts of user utterances that need to be clustered according to their intent. In these commercial settings, no labeled data is typically available, and the number of clusters is not known. Our method is based on iterative vector updating: it constructs sparse vectors based on representative texts, and then iteratively refines them through LLM guidance. Our method achieves comparable or superior results to state-of-the-art methods that use contrastive learning, but without assuming prior knowledge of clusters or labels. Experiments on diverse datasets and smaller LLMs show that our method is model agnostic and can be applied to any embedder, with relatively small LLMs, and different clustering methods. We also show that our method scales to large datasets, reducing the computational cost of the LLM. These low-resource, adaptable settings and the scalability of our method make it more aligned with real-world scenarios than existing clustering methods.