Research map / Learning theory and algorithms / Combinatorial and search optimization
Learning to Segment for Vehicle Routing Problems
Wenbin Ouyang, Sirui Li, Yining Ma, Cathy Wu
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add L2Seg-SYN segmentation to the LNS backbone (ablation) | LNS · CVRP2k | objective: 44.92 -> 43.42 | Improved |
| add L2Seg-SYN segmentation to the LNS backbone (ablation) | LNS · CVRP5k | objective: 64.69 -> 63.94 | Improved |
| add enhanced features to L2Seg-SYN (ablation) | L2Seg-SYN-LNS · CVRP2k | objective: 43.65 -> 43.42 | Improved |
| add enhanced features to L2Seg-SYN (ablation) | L2Seg-SYN-LNS · CVRP5k | objective: 64.22 -> 63.94 | Improved |
| replace learned unstable-edge detection with random 40% edges in FSTA (ablation) | LNS · CVRP2k | objective: 44.92 -> 46.24 | Hurt |
| replace learned unstable-edge detection with random 60% edges in FSTA (ablation) | LNS · CVRP5k | objective: 64.69 -> 65.92 | Hurt |
| apply L2Seg-SYN to the LKH-3 backbone | LKH-3 · CVRP5k | objective: 175.70 -> 122.34 | Improved |
| apply L2Seg-SYN to the LNS backbone | LNS · CVRP2k | objective: 57.62 -> 56.08 | Improved |
| apply L2Seg-SYN to the L2D backbone | L2D · VRPTW5k | objective: 335.2 -> 333.4 | Improved |
| apply L2Seg-SYN to the LKH-3 backbone | LKH-3 · VRPTW5k | objective: 353.2 -> 345.2 | Improved |
| use a perfect-recall-and-TNR oracle for unstable edges instead of the learned model | LNS · CVRP2k | time to reach L2Seg-SYN-LNS quality: 241s -> 39s | Improved |
| use an oracle with 70% recall and TNR instead of the learned model | LNS · CVRP2k | time to reach L2Seg-SYN-LNS quality: 241s -> 324s | Hurt |
Closest papers by meaning
- MVMoE: Multi-Task Vehicle Routing Solver with Mixture-of-Experts (ICML 2024 · similarity 0.89)
- Multi-Task Vehicle Routing Solver via Mixture of Specialized Experts under State-Decomposable MDP (NeurIPS 2025 · similarity 0.89)
- Improving Generalization of Neural Combinatorial Optimization for Vehicle Routing Problems via Test-Time Projection Learning (NeurIPS 2025 · similarity 0.88)
- SPACE: Unifying Symmetric and Asymmetric Routing Problems for Generalist Neural Solver (NeurIPS 2026 · similarity 0.88)
- CaDA: Cross-Problem Routing Solver with Constraint-Aware Dual-Attention (ICML 2025 · similarity 0.88)
- A Mixed-Curvature based Pre-training Paradigm for Multi-Task Vehicle Routing Solver (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.
