#machine learning
Mar 2026
Generating from Discrete Distributions Using Diffusions: Insights from Random Constraint Satisfaction Problems
This paper studies the problem of generating a uniformly random solution of a given (random) $k$-SAT or $k$-XORSAT formula and observes that continuous diffusions and learned diffusions can match the theoretical `ideal' accuracy.
Alankrita Bhatt, Mukur Gupta, Germain Kolossov et al.
· arXiv.org · 2 citations