Research map / Graph learning / Graph algorithms and clustering
OptMaster: A DAG-Based Framework for Formulation and Heuristic Discovery in Optimization
Hang Lin, Yuanpeng Gao, Yuzhi Zhang, Kun Yuan, Gang Yan, Siheng Chen, Linfeng Zhang, Weinan E
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 Code as Verifier to node scoring (Base) (ablation) | OptMaster · IndustryOR | accuracy (%): 75.0 -> 82.7 | Improved |
| add Code as Verifier to node scoring (Two-Interpretations) (ablation) | OptMaster · IndustryOR | accuracy (%): 77.7 -> 87.3 | Improved |
| add Code as Verifier to node scoring (Two-Interpretations) (ablation) | OptMaster · Mamo-Complex | accuracy (%): 94.6 -> 95.6 | Improved |
| add Code as Verifier to node scoring (Base) (ablation) | OptMaster · OptMATH | accuracy (%): 53.2 -> 57.6 | Improved |
| add Code as Verifier (ablation) | OptMaster · MAXCUT Gset70 | cut size: 9529 -> 9587 | Improved |
| add Code as Verifier (ablation) | OptMaster · Circle Packing n=26 | sum of radii: 2.62840584 -> 2.63598308 | Improved |
| use multi-parent DAG (k=3) instead of tree (k=1) or chain (ablation) | OptMaster · MAXCUT Gset70 and Circle Packing n=26 | best objective | Improved |
| add Two-Interpretations exploration planning on top of Base (ablation) | OptMaster · Mamo-Complex | accuracy (%): 80.1 -> 95.6 | Improved |
| add OptMaster framework to GPT-5 (Base) | GPT-5 · IndustryOR | accuracy (%): 72.0 -> 82.7 | Improved |
| add OptMaster discovery loop to GPT-5 | GPT-5 · MAXCUT Gset70 | cut size: 9379 -> 9587 | Improved |
| extend runtime budget of discovered heuristic to 6 hours (ablation) | OptMaster BPGD heuristic · MAXCUT Gset70 | cut size: 9587 -> 9590 | Improved |
Closest papers by meaning
- Step-by-Step Optimization-like Reasoning in LLMs over Expanding Search Spaces (NeurIPS 2026 · similarity 0.88)
- OptiBench Meets ReSocratic: Measure and Improve LLMs for Optimization Modeling (ICLR 2025 · similarity 0.88)
- OptiTree: Hierarchical Thoughts Generation with Tree Search for LLM Optimization Modeling (NeurIPS 2025 · similarity 0.87)
- MM-OptBench: A Solver-Grounded Benchmark for Multimodal Optimization Modeling (NeurIPS 2026 · similarity 0.86)
- PathWise: Planning through World Model for Automated Heuristic Design via Self-Evolving LLMs (ICML 2026 · similarity 0.86)
- Autoformulation of Mathematical Optimization Models Using LLMs (ICML 2025 · similarity 0.86)
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.
