AI Paper Atlas

Research map / Graph learning / Graph algorithms and clustering

CoCo-MILP: Inter-Variable Contrastive and Intra-Constraint Competitive MILP Solution Prediction

Tianle Pu, Jianing Li, Yingying Gao, Shixuan Liu, Zijie Geng, Haoyang Liu, Chao Chen, Changjun Fan

AAAI 2026 · Graph learning › Graph algorithms and clustering · approach: Neural combinatorial optimization on graphs · 1 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 CoCo-MILP (VCL + ICC layer) to Gurobi versus plain GurobiGurobi · SC, 1000sabsolute primal gap: 0.16 -> 0.01Improved
add CoCo-MILP to GurobiGurobi · CA, 1000sabsolute primal gap: 295.44 -> 185.73Improved
CoCo-MILP versus ConPSGurobi · CA, 1000sabsolute primal gap: 208.54 -> 185.73Improved
add PS or ConPS to Gurobi on WAGurobi · WA, 1000sabsolute primal gap: 0.42 -> 0.42No clear effect
replace BCE with Inter-Variable Contrastive Loss (VCL) (ablation)CoCo-MILP · SC, 1000sobjective (minimize): 125.25 -> 125.06Improved
replace BCE with VCL (ablation)CoCo-MILP · CA, 1000sobjective (maximize): 97240.00 -> 97338.64Improved
add pairwise ranking loss to VCL (ablation)CoCo-MILP · CA, 1000sobjective (maximize): 97272.41 -> 97338.64Improved
add multi-sample contrastive loss to VCL (ablation)CoCo-MILP · CA, 1000sobjective (maximize): 97217.40 -> 97338.64Improved
add Intra-Constraint Competitive layer to the GNN (ablation)CoCo-MILP · SC, 1000sobjective (minimize): 125.18 -> 125.06Improved
add ICC layer to the GNN (ablation)CoCo-MILP · CA, 1000sobjective (maximize): 97315.96 -> 97338.64Improved
CoCo-MILP versus Gurobi on MIPLIB IIS instancesGurobi · MIPLIB IIS subsetbest objective: fast0507 174.00 -> 174.00; ramos3 233.00 -> 224.00Mixed

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