Metraipolis › archive › dc_e8a19dd1da64
Self-Evolving Algorithm-Design Agents: Escaping In-Context Evolutionary Stagnation via Population-Curated Policy Optimization — arXiv:2609.38757
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38757
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T12:42:08.131Z
- SHA-256 of the archived text
ad02a6adbb3d86a634de5732a8e4c9a5d81a9c6d9fa6b0380d4e14ab6eb62964- Stable URL of this page
https://metraipolis.com/documento/dc_e8a19dd1da64
How to cite
Self-Evolving Algorithm-Design Agents: Escaping In-Context Evolutionary Stagnation via Population-Curated Policy Optimization — arXiv:2609.38757 — arXiv (en), https://arxiv.org/abs/2609.38757 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e8a19dd1da64 (Metraipolis archive dc_e8a19dd1da64, sha256 ad02a6adbb3d86a634de5732a8e4c9a5d81a9c6d9fa6b0380d4e14ab6eb62964)
BibTeX
@misc{dce8a19dd1da64,
title = {Self-Evolving Algorithm-Design Agents: Escaping In-Context Evolutionary Stagnation via Population-Curated Policy Optimization — arXiv:2609.38757},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38757},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e8a19dd1da64; sha256 ad02a6adbb3d86a634de5732a8e4c9a5d81a9c6d9fa6b0380d4e14ab6eb62964},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38757. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models are increasingly participating in complex real-world tasks in the form of algorithm-design agents, designing and refining algorithms. Many successful algorithm-design agents adopt pure in-context evolutionary frameworks, but they may quickly plateau in domains that require specialized knowledge. Parametric adaptation offers a way to internalize specialized knowledge, but conventional training requires abundant domain-specific corpora while high-quality algorithms are scarce in complex algorithm-design scenarios. In this paper, we propose sample-efficient parametric self-evolution where agents can explore and learn from self-generated algorithms. First, we characterize in-context evolutionary stagnation and analytically propose the Improvement Chain proposition, showing how learning successive self-generated algorithms can locally increase the likelihood of neighboring algorithms. Motivated by this local-transfer perspective, we further propose Population-Curated Policy Optimization (PCPO) to utilize a global population and a hybrid policy update scheme for retaining and reusing high-quality, diverse self-generated algorithms, shifting the policy towards stronger algorithms. In the task of learning rate schedule design for global placement in electronic design automation, trained only on 4 chip cases, PCPO outperforms the state-of-the-art in-context evolutionary methods (e.g., OpenEvolve and ShinkaEvolve) on average across 16 chip cases. With an 8B-size base model, PCPO achieves competitive performance compared to frontier closed-source models such as GPT-5.5. PCPO also reduces inference-time token cost by internalizing grounded domain knowledge and prompt distillation. Moreover, PCPO achieves significant speedups on four GPU kernel designs, with an average of 8.27$\times$ speedup against the PyTorch Eager baseline. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Neural and Evolutionary Computing (cs.NE). Pagina del lavoro: https://arxiv.org/abs/2609.38757 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T12:14:13.677Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).