4 papers
Optimal Recovery Meets Bayesian Learning: Where Worst-Case Bounds Pay Off
Gordei Verbii
Worst-case Optimal Recovery (OR) and Bayesian learning describe the same Gaussian-quadratic-Hilbert problems in two vocabularies. We sharpen the correspondence - the radius of info…
One Inverse Step is a Convex Program: Bayes-Limit Calibration of Diffusion Inversion
Gordei Verbii
One implicit DDIM inversion step is the cheapest probe of whether a pretrained diffusion model encodes local manifold geometry. It is the stationarity condition of an explicit pote…
The Distributional View of Knowledge Distillation
Gordei Verbii, Juho Lee
Token-level knowledge distillation (KD) matches two conditional distributions per position, yet the standard objectives compare them pointwise: a Kullback-Leibler gradient is blind…
CayleyPy RL: Pathfinding and Reinforcement Learning on Cayley Graphs
A. Chervov, M. Obozov, A. Soibelman +31
This paper is the second in a series of studies on developing efficient artificial intelligence-based approaches to pathfinding on extremely large graphs (e.g. nodes) wit…