From Topology to Retrieval: Decoding Embedding Spaces with Unified Signatures
By: Florian Rottach , William Rudman , Bastain Rieck and more
Potential Business Impact:
Maps text meaning to help computers find information.
Studying how embeddings are organized in space not only enhances model interpretability but also uncovers factors that drive downstream task performance. In this paper, we present a comprehensive analysis of topological and geometric measures across a wide set of text embedding models and datasets. We find a high degree of redundancy among these measures and observe that individual metrics often fail to sufficiently differentiate embedding spaces. Building on these insights, we introduce Unified Topological Signatures (UTS), a holistic framework for characterizing embedding spaces. We show that UTS can predict model-specific properties and reveal similarities driven by model architecture. Further, we demonstrate the utility of our method by linking topological structure to ranking effectiveness and accurately predicting document retrievability. We find that a holistic, multi-attribute perspective is essential to understanding and leveraging the geometry of text embeddings.
Similar Papers
From Topology to Retrieval: Decoding Embedding Spaces with Unified Signatures
Machine Learning (CS)
Maps text meaning to help computers find information.
Harnessing the Universal Geometry of Embeddings
Machine Learning (CS)
Lets computers understand different word meanings.
Representation Integrity in Temporal Graph Learning Methods
Machine Learning (CS)
Checks if computer network changes are understood.