Guided Discrete Diffusion for Constraint Satisfaction Problems
On the map: Diffusion models → Discrete and text diffusion
Abstract (arXiv)
We propose discrete diffusion guidance for constraint satisfaction problems (CSPs) and demonstrate its ability to solve Sudoku puzzles without supervision.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 9 of its 9 references that are in the map. First posted on arXiv 16 Dec 2025.
Builds on
9 of 9 references.
Closest ancestors · 1
Same direction, newest first
Other references · 8
- 2016Categorical Reparameterization with Gumbel-Softmaxcited by 2,708
- 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
- 2019Plug and Play Language Models: A Simple Approach to Controlled Text Generationcited by 793
- 2022A Continuous Time Framework for Discrete Denoising Modelscited by 437
- 2022Understanding Diffusion Models: A Unified Perspectivecited by 337
- 2023Protein Design with Guided Discrete Diffusioncited by 154
- 2017Recurrent Relational Networkscited by 71
- 2023Learning to Solve Constraint Satisfaction Problems with Recurrent Transformercited by 21
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2026Can Continuous-Time Diffusion Models Generate and Solve Globally Constrained Discrete Problems? A Study on Sudoku
- 2026DiBS: Diffusion-Informed Branch Selection
- 2023Compositional Diffusion-Based Continuous Constraint Solvers
- 2025Teaching Transformers to Solve Combinatorial Problems through Efficient Trial & Error
- 2026Generating from Discrete Distributions Using Diffusions: Insights from Random Constraint Satisfaction Problems
- 2026Unsupervised Diffusion Solver for Combinatorial Optimization via Combinatorial Adjoint Matching
- 2026Diffusion as a Training Curriculum for Timestep-Free Iterative Reasoning
- 2025Constraints-Guided Diffusion Reasoner for Neuro-Symbolic Learning