CDEoH: Category-Driven Automatic Algorithm Design With Large Language Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911530807721984 |
|---|---|
| author | Wang, Yu-Nian Lyu, Shen-Huan Chen, Ning Xu, Jia-Le Ye, Baoliu Zhang, Qingfu |
| author_facet | Wang, Yu-Nian Lyu, Shen-Huan Chen, Ning Xu, Jia-Le Ye, Baoliu Zhang, Qingfu |
| contents | With the rapid advancement of large language models (LLMs), LLM-based heuristic search methods have demonstrated strong capabilities in automated algorithm generation. However, their evolutionary processes often suffer from instability and premature convergence. Existing approaches mainly address this issue through prompt engineering or by jointly evolving thought and code, while largely overlooking the critical role of algorithmic category diversity in maintaining evolutionary stability. To this end, we propose Category Driven Automatic Algorithm Design with Large Language Models (CDEoH), which explicitly models algorithm categories and jointly balances performance and category diversity in population management, enabling parallel exploration across multiple algorithmic paradigms. Extensive experiments on representative combinatorial optimization problems across multiple scales demonstrate that CDEoH effectively mitigates convergence toward a single evolutionary direction, significantly enhancing evolutionary stability and achieving consistently superior average performance across tasks and scales. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_19284 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | CDEoH: Category-Driven Automatic Algorithm Design With Large Language Models Wang, Yu-Nian Lyu, Shen-Huan Chen, Ning Xu, Jia-Le Ye, Baoliu Zhang, Qingfu Neural and Evolutionary Computing Artificial Intelligence With the rapid advancement of large language models (LLMs), LLM-based heuristic search methods have demonstrated strong capabilities in automated algorithm generation. However, their evolutionary processes often suffer from instability and premature convergence. Existing approaches mainly address this issue through prompt engineering or by jointly evolving thought and code, while largely overlooking the critical role of algorithmic category diversity in maintaining evolutionary stability. To this end, we propose Category Driven Automatic Algorithm Design with Large Language Models (CDEoH), which explicitly models algorithm categories and jointly balances performance and category diversity in population management, enabling parallel exploration across multiple algorithmic paradigms. Extensive experiments on representative combinatorial optimization problems across multiple scales demonstrate that CDEoH effectively mitigates convergence toward a single evolutionary direction, significantly enhancing evolutionary stability and achieving consistently superior average performance across tasks and scales. |
| title | CDEoH: Category-Driven Automatic Algorithm Design With Large Language Models |
| topic | Neural and Evolutionary Computing Artificial Intelligence |
| url | https://arxiv.org/abs/2603.19284 |