Nonparametric In-Context Learning under Growing Geometric Complexity: Minimax Optimality and Local Geometry-Adaptivity of Transformers
stat.ML, cs.LG, math.ST, stat.TH
Submitted: 2026-09-25
Updated: 2026-09-25
Code: https://github.com/seojaehee02/nonparametric_ICL
Terminology
Sources
- Unraveling the Localized Latents: Learning Stratified Manifold Structures in LLM Embedding Space with Sparse Mixture-of-Experts
- A Deep Generative Approach to Stratified Learning
- Transformers for Learning on Noisy and Task-Level Manifolds: Approximation and Generalization Insights
Related papers
- Behavior of prediction performance metrics with rare events
- Optimal Estimation of Generic Dynamics by Path-Dependent Neural Jump ODEs
- A Posterior-Dynamics Framework for Imaging Inverse Problems with Pretrained Diffusion Priors
- Efficient quadratic entropy with distance sketches
- Embedding-Bias in Conditional Independence Testing
- From Log-Odds to Shapley Values: An Explanatory Geometry for the Weighted Naive Bayes Classifier