Demystifying Embedding Spaces using Large Language Models
This addresses the challenge of interpreting embeddings for users in fields like machine learning and data analysis, but it is incremental as it builds on existing methods by integrating LLMs.
The paper tackled the problem of making embeddings more interpretable by using Large Language Models to transform abstract vectors into understandable narratives, enabling querying and exploration of complex embedding data across tasks like concept activation vectors and recommender systems.
Embeddings have become a pivotal means to represent complex, multi-faceted information about entities, concepts, and relationships in a condensed and useful format. Nevertheless, they often preclude direct interpretation. While downstream tasks make use of these compressed representations, meaningful interpretation usually requires visualization using dimensionality reduction or specialized machine learning interpretability methods. This paper addresses the challenge of making such embeddings more interpretable and broadly useful, by employing Large Language Models (LLMs) to directly interact with embeddings -- transforming abstract vectors into understandable narratives. By injecting embeddings into LLMs, we enable querying and exploration of complex embedding data. We demonstrate our approach on a variety of diverse tasks, including: enhancing concept activation vectors (CAVs), communicating novel embedded entities, and decoding user preferences in recommender systems. Our work couples the immense information potential of embeddings with the interpretative power of LLMs.