Embedding Comparator: Visualizing Differences in Global Structure and Local Neighborhoods via Small Multiples
This addresses the problem of inefficient embedding comparison for users across disciplines, though it is incremental as it builds on existing visualization techniques.
The paper tackles the tedious task of comparing embeddings by developing the Embedding Comparator, an interactive system that visualizes global structure and local neighborhoods, which in evaluations with 15 participants accelerated comparisons by shifting from manual specification to browsing visualizations.
Embeddings mapping high-dimensional discrete input to lower-dimensional continuous vector spaces have been widely adopted in machine learning applications as a way to capture domain semantics. Interviewing 13 embedding users across disciplines, we find comparing embeddings is a key task for deployment or downstream analysis but unfolds in a tedious fashion that poorly supports systematic exploration. In response, we present the Embedding Comparator, an interactive system that presents a global comparison of embedding spaces alongside fine-grained inspection of local neighborhoods. It systematically surfaces points of comparison by computing the similarity of the $k$-nearest neighbors of every embedded object between a pair of spaces. Through case studies across multiple modalities, we demonstrate our system rapidly reveals insights, such as semantic changes following fine-tuning, language changes over time, and differences between seemingly similar models. In evaluations with 15 participants, we find our system accelerates comparisons by shifting from laborious manual specification to browsing and manipulating visualizations.