Metraipolis › archive › dc_72199472e97d

Relay, Don't Route: Adaptive Population Handoff for Cost-Efficient LLM-Driven Evolution — arXiv:2608.05651

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.05651
License (as archived)
CC0-1.0
Archived
2026-10-31T10:50:09.523Z
SHA-256 of the archived text
7dcbec3e8cc599f18c4453588d8b6a886599cd0624a252a1ffefa715053c99c7
Stable URL of this page
https://metraipolis.com/documento/dc_72199472e97d

How to cite

Relay, Don't Route: Adaptive Population Handoff for Cost-Efficient LLM-Driven Evolution — arXiv:2608.05651 — arXiv (en), https://arxiv.org/abs/2608.05651 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_72199472e97d (Metraipolis archive dc_72199472e97d, sha256 7dcbec3e8cc599f18c4453588d8b6a886599cd0624a252a1ffefa715053c99c7)
BibTeX
@misc{dc72199472e97d,
  title = {Relay, Don't Route: Adaptive Population Handoff for Cost-Efficient LLM-Driven Evolution — arXiv:2608.05651},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.05651},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_72199472e97d; sha256 7dcbec3e8cc599f18c4453588d8b6a886599cd0624a252a1ffefa715053c99c7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.05651. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large language model (LLM)-driven evolution has shown promise for program search and algorithm discovery, but relying on strong models throughout long evolutionary runs is costly. A natural alternative is to combine cheap and strong models under a fixed inference budget. However, existing approaches typically allocate models at the level of individual queries or mutation steps, overlooking that evolutionary search is \textit{stateful}: each generated candidate changes the population from which subsequent mutations are produced. We empirically analyze LLM-driven evolutionary trajectories and find that search progress is strongly front-loaded, early trajectory performance is informative but noisy, and cheap models recover much of the early progress achieved by strong models at lower cost. Motivated by these findings, we propose \textbf{\model}, a training-free framework that shifts budget allocation from individual calls to evolving populations through adaptive \textit{population handoff}. A cheap model explores multiple trajectories in short blocks allocated by a bandit scheduler. Relay Gain, defined as the marginal improvement of a compact, quality-diverse candidate bank constructed for handoff, serves as the scheduler reward and determines when to hand off. The curated candidates initialize a shared strong model population for refinement. Across four benchmarks and three budgets, \model achieves the highest mean score in 11 of 12 settings, outperforming competitive baselines. Our results suggest that in stateful search, budget allocation should be organized around the population, not the individual call.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-07 in cs.NE).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE).
Pagina del lavoro: https://arxiv.org/abs/2608.05651 — 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 2026-10-31T10:36:29.960Z dalla pagina https://arxiv.org/catchup/cs.NE/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).