Siamak Ravanbakhsh

The Expressive Limits of Diagonal SSMs for State-Tracking

Inverting Data Transformations via Diffusion Sampling

Long-Horizon Model-Based Offline Reinforcement Learning Without Conservatism

Scaling Laws and Symmetry, Evidence from Neural Force Fields

Progressive Inference-Time Annealing of Diffusion Models for Sampling from Boltzmann Densities

Beyond Scalar Rewards: An Axiomatic Framework for Lexicographic MDPs

SymmCD: Symmetry-Preserving Crystal Generation with Diffusion Models

Energy Loss Functions for Physical Systems

Diffusion Tree Sampling: Scalable inference‑time alignment of diffusion models

Sampling from Energy-based Policies using Diffusion