Research map / Image generation / Fast diffusion sampling
Generation as Search Operator for Test-Time Scaling of Diffusion-based Combinatorial Optimization
Yang Li, Lvda Chen, Haonan Wang, Runzhong Wang, Junchi Yan
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes GenSCO, which scales diffusion-based combinatorial optimization solvers at inference time by using generation as a search operator, an axis orthogonal to adding more denoising steps.
Problem
Diffusion solvers for combinatorial optimization improve solution quality mainly by adding denoising steps, but their performance saturates quickly, and their inference-time cost-efficiency is underexplored.
Approach
Treats the diffusion generation process itself as a search operator and scales inference computation along a dimension orthogonal to the number of denoising steps, rather than only increasing denoising steps.
Closest papers by meaning
- Fast T2T: Optimization Consistency Speeds Up Diffusion-Based Training-to-Testing Solving for Combinatorial Optimization (NeurIPS 2024 · similarity 0.91)
- Efficient Combinatorial Optimization via Heat Diffusion (NeurIPS 2024 · similarity 0.88)
- A Unified Sampling Framework for Solver Searching of Diffusion Probabilistic Models (ICLR 2024 · similarity 0.88)
- TFG: Unified Training-Free Guidance for Diffusion Models (NeurIPS 2024 · similarity 0.87)
- Test-time Alignment of Diffusion Models without Reward Over-optimization (ICLR 2025 · similarity 0.87)
- A First-order Generative Bilevel Optimization Framework for Diffusion Models (ICML 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
