Consistent estimation of generative model representations in the data kernel perspective space
This work addresses the need for techniques to study evolving generative models, though it appears incremental as it builds on existing embedding-based representations.
The paper tackles the problem of analyzing differences in behavior among generative models by establishing theoretical conditions for consistently estimating model embeddings as the number of queries and models increases.
Generative models, such as large language models and text-to-image diffusion models, produce relevant information when presented a query. Different models may produce different information when presented the same query. As the landscape of generative models evolves, it is important to develop techniques to study and analyze differences in model behaviour. In this paper we present novel theoretical results for embedding-based representations of generative models in the context of a set of queries. In particular, we establish sufficient conditions for the consistent estimation of the model embeddings in situations where the query set and the number of models grow.