From Distribution Learning in Training to Gradient Search in Testing for Combinatorial Optimization

Yang Li, Jinpei Guo, Runzhong Wang, Junchi Yan
2/12/2026

No official code found. Best implementation: Thinklab-SJTU/T2TCO (73 stars, MIT, last commit 2025-07-02). 1 more below.

DOISemantic Scholar

Code Implementations(2)

Thinklab-SJTU/T2TCOCommunity75%
MIT

[NeurIPS 2023] T2T: From Distribution Learning in Training to Gradient Search in Testing for Combinatorial Optimization

738Shell, CythonNov 1, 20231 years agoMIT
combinatorial-optimizationdiffusiongenerative-modelgraphmachine-learning
Thinklab-SJTU/Fast-T2TCommunity62%
MIT

[NeurIPS2024] Fast T2T: Optimization Consistency Speeds Up Diffusion-Based Training-to-Testing Solving for Combinatorial Optimization; [NeurIPS2023] T2T: From Distribution Learning in Training to Gradient Search in Testing for Combinatorial Optimization

211C, C++Dec 9, 20241 years agoMIT

Cite this paper

@article{li2026from,
  title  = {From Distribution Learning in Training to Gradient Search in Testing for Combinatorial Optimization},
  author = {Yang Li and Jinpei Guo and Runzhong Wang and Junchi Yan},
  year   = {2026},
  doi    = {10.52202/075280-2176},
  url    = {https://doi.org/10.52202/075280-2176},
  journal = {NEURIPS 2023 2023}
}

Discussion