KL Before and After Mixing: A Round-Complexity Separation in Parallel Sampling
cs.LG, cs.IT, math.IT
Submitted: 2026-10-09
Updated: 2026-10-09
Comments: 15 pages of main text; 169 pages total including references and appendices; 7 figures. Formal proof materials included as ancillary files
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Quadratic Speedups in Parallel Sampling from Determinantal Distributions
- Parallel Sampling via Autospeculation
- Walk fast but be careful: Understanding Parallel Sampling in Masked Diffusion
- Metric Diophantine Approximation: aspects of recent work
- Approximation of rejective sampling inclusion probabilities and application to high order correlations
- Confidence-Based Decoding is Provably Efficient for Diffusion Language Models
- Parallelism, critical windows, and separations among diffusion language models
- Optimal Inference Schedules for Masked Diffusion Models
- Efficient Sampling with Discrete Diffusion Models: Sharp and Adaptive Guarantees
- Theoretical Benefit and Limitation of Diffusion Language Model
- Learning Generation Orders for Masked Discrete Diffusion Models via Variational Inference
- Batched Multi-armed Bandits Problem
- Autoregressive Diffusion Models
- Counterfactual Probing for Parallel Unmasking with Hidden Forest Structure
- Lower Bounds for Parallel Diffusion Sampling
- Error Bounds and Optimal Schedules for Masked Diffusions with Factorized Approximations
- Breaking AR's Sampling Bottleneck: Provable Acceleration via Diffusion Language Models
- PyTorch: An Imperative Style, High-Performance Deep Learning Library
- Schedule optimization for tau-leaping in masked discrete diffusion
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks