Research map / Large language models / Agents, code and math
Hierarchical Representations for Cross-task Automated Heuristic Design using LLMs
Fei Liu, Rui Zhang, Shunyu Yao, Qinglong Hu, kefeng zheng, Zhichao Lu, Qingfu Zhang
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 multi-task knowledge transfer and Pareto population management (MTHS vs STHS) (ablation) | MTHS · CVRPLib set X | gap to best-known (%): 5.36 -> 4.29 | Improved |
| add multi-task knowledge transfer (MTHS vs STHS) (ablation) | MTHS · TSPLib 500-1000 | gap to best-known (%): 3.59 -> 2.64 | Improved |
| add multi-task knowledge transfer (MTHS vs STHS) (ablation) | MTHS · TSPLib 50-99 | gap to best-known (%): 0.87 -> 0.72 | Mixed |
| use hierarchical MTHS evolved metaheuristic vs single-task LLM AHD baselines on FSSP | FSSP heuristic design · Taillard FSSP average | gap to upper bound (%): 0.68 (ILS) -> 0.24 | Improved |
| use structured template metaheuristic representation instead of abstract, pseudocode or thought (ablation) | MTHS · CVRP and FSSP convergence | best score vs samples | Improved |
| guide LLM with the MTHS metaheuristic (+MH) on unseen BPP | Gemini-2.5-pro · BPP n500, c150 | gap to lower bound: 1.32% -> 0.34% | Improved |
| guide GPT-5-mini with the MTHS metaheuristic on unseen BPP | GPT-5-mini · BPP n1000, c150 | gap to lower bound: 0.98% -> 0.65% | Improved |
| guide LLM with the MTHS metaheuristic on black-box optimization | GPT-5-mini · BBOP average of top-10 | score (lower is better): 2.86 -> 0.32 | Improved |
| guide GPT-5-mini with the MTHS metaheuristic on admissible set problem | GPT-5-mini · ASP average of top-10 | score (lower is better): 631.5 -> 899.1 | Hurt |
| guide Gemini-2.5-pro with the MTHS metaheuristic on admissible set problem | Gemini-2.5-pro · ASP average of top-10 | score (lower is better): 824.1 -> 742.6 | Mixed |
| multi-task search over three tasks instead of single-task searches | MTHS · TSP, CVRP, FSSP | number of code evaluations: about 67% fewer evaluations | Improved |
Closest papers by meaning
- HSEvo: Elevating Automatic Heuristic Design with Diversity-Driven Harmony Search and Genetic Algorithm Using LLMs (AAAI 2025 · similarity 0.90)
- Multi-Objective Evolution of Heuristic Using Large Language Model (AAAI 2025 · similarity 0.89)
- PathWise: Planning through World Model for Automated Heuristic Design via Self-Evolving LLMs (ICML 2026 · similarity 0.88)
- Monte Carlo Tree Search for Comprehensive Exploration in LLM-Based Automatic Heuristic Design (ICML 2025 · similarity 0.88)
- Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Model (ICML 2024 · similarity 0.88)
- ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution (NeurIPS 2024 · 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.
