AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Unsupervised Diffusion Solver for Combinatorial Optimization via Combinatorial Adjoint Matching

Shengyu Feng, Tarun Suresh, Yiming Yang

ICML 2026 · Image generation › Fast diffusion sampling · approach: Diffusion theory and learned samplers · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add adjoint propagation to the TSP training target (ablation)CAM · TSP-500optimality gap: 2.83% -> 2.18%Improved
use 2-OPT instead of greedy decoding for the terminal heuristic (ablation)CAM · TSP-500optimality gap: 2.36% -> 2.18%Improved
use CAM instead of SDDS as unsupervised training objectivediffusion CO solver · MIS ER-[9000-11000]independent set size: 350.63 -> 384.38Improved
use CAM instead of RLNNdiffusion CO solver · MIS RB-[800-1200]independent set size: 39.96 -> 41.25Improved
use CAM instead of DiffUCOdiffusion CO solver · MIS RB-[800-1200]independent set size: 40.52 -> 41.25Improved
use CAM instead of SDDSdiffusion CO solver · TSP-1000tour length: 23.90 -> 23.70Improved
use CAM instead of SDDSdiffusion CO solver · CVRP-100route length: 16.16 -> 15.98Improved
use CAM instead of supervised DIFUSCOdiffusion CO solver · TSP-500tour length: 16.85 -> 16.91Hurt
use CAM instead of supervised DIFUSCOdiffusion CO solver · CVRP-100route length: 15.97 -> 15.98Hurt
use CAM instead of SDDSdiffusion CO solver · Max Cut BA-[800-1200]cut size: 2963.95 -> 2964.20No clear effect
use CAM instead of the OR solver MQLibMax Cut solver · Max Cut BA-[800-1200]cut size: 2961.38 -> 2964.20Improved
use CAM instead of the OR solver KaMISMIS solver · MIS ER-[9000-11000] (transfer)independent set size: 381.31 -> 384.38Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy