Scott Linderman

An Information Theoretic Perspective on Agentic System Design

A cross-species neural foundation model for end-to-end speech decoding

SING: SDE Inference via Natural Gradients

Predictability Enables Parallelization of Nonlinear State Space Models

Extracting task-relevant preserved dynamics from contrastive aligned neural recordings

Identifying multi-compartment Hodgkin-Huxley models with high-density extracellular voltage recordings

Parallelizing MCMC Across the Sequence Length

Cost-efficient Collaboration between On-device and Cloud Language Models

Informed Correctors for Discrete Diffusion Models

Weaver: Shrinking the Generation-Verification Gap by Scaling Compute for Verification

On the interplay between learning and memory in deep state space models