Smoothed Score Queries and the Complexity of Sampling
cs.DS, cs.IT, cs.LG, math.IT, stat.ML
Submitted: 2026-05-26
Updated: 2026-11-07
Terminology
Sources
- High-accuracy sampling for diffusion models and log-concave distributions
- Sampling is as easy as learning the score: theory for diffusion models with minimal data assumptions
- Query lower bounds for log-concave sampling
- Fisher information lower bounds for sampling
- Convergence Analysis for General Probability Flow ODEs of Diffusion Models in Wasserstein Distances
- Instance-dependent Convergence Theory for Diffusion Models
- Optimal Convergence Analysis of DDPM for General Distributions
- Fast and Efficient Parallel Sampling Using Higher Order Langevin Dynamics
- Query Lower Bounds for Diffusion Sampling
- Sublinear iterations can suffice even for DDPMs
Related papers
- PHastR: Faster Minimal Perfect Hashing with Rotated Layouts
- Fine-Grained Complexity of Approximating Vector Knapsack: A Faster Algorithm and Equivalence with k-SUM in 2D
- Polynomial-time computation of p-contraction fixed points for even p
- Separations Between Submodular-Like Graph Widths
- Bounded Integer Programming in Almost N 2d Time
- Fast Odd-Permutation Sums in Characteristic Two and Shortest Even Directed Cycles