Tag
This paper proves that task-relevant latent representations can be identified from generalist models in a fully nonparametric setting without interventions or parametric constraints, achieving a hierarchical identifiability guarantee across time steps and within each step.
This paper establishes nonparametric identifiability guarantees for extracting task-relevant representations from generalist models, proving that task structure is identifiable across time steps and latent representations are identifiable within each step under sparsity regularization.
The paper introduces diverse dictionary learning, showing that key set-theoretic relationships among latent variables can be identified from observational data without strong assumptions, enabling partial or full identifiability with minimal inductive bias.