Metraipolis › archive › dc_56154d86cf56
The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance Rules — arXiv:2610.09239
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.09239
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T11:09:09.590Z
- SHA-256 of the archived text
cf92db1fcb37c5320aac3c6f4164550d9637c696616b35703ba65c641e98357d- Stable URL of this page
https://metraipolis.com/documento/dc_56154d86cf56
How to cite
The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance Rules — arXiv:2610.09239 — arXiv (en), https://arxiv.org/abs/2610.09239 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_56154d86cf56 (Metraipolis archive dc_56154d86cf56, sha256 cf92db1fcb37c5320aac3c6f4164550d9637c696616b35703ba65c641e98357d)
BibTeX
@misc{dc56154d86cf56,
title = {The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance Rules — arXiv:2610.09239},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.09239},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_56154d86cf56; sha256 cf92db1fcb37c5320aac3c6f4164550d9637c696616b35703ba65c641e98357d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09239. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Self-improving LLM systems propose changes to themselves and keep those that score better on a small evaluation set. We treat this keep-if-better step as selection under measurement noise, model the correlated errors of the candidates in a single decision, and study empirically what happens when the evaluation set is reused. In runs where Qwen models rewrite their own instructions and every candidate is also scored on 600 held-out items, most proposals after the first are harmful, and the model gives the size of the winner's curse of a generation's best candidate. With a prior from a separate pilot, it matches the average overstatement of first-generation commits in native loops, though not setting by setting. In a pre-registered study, the final selection-set score of greedy loops exceeded held-out accuracy by 13 to 20 points with 16 selection items and by 1 to 5 points with 256. Held-out gains grew with the selection set on TREC but not on GSM8K, and the tested acceptance rules did not beat greedy acceptance over whole runs. Gains measured on the selection set also exceeded held-out gains when a current model refined a competent instruction, and in the validation scores of GEPA and MIPROv2. Scoring the starting and the current instruction on 64 items never used for selection removes the average bias of a loop's reported gain, but single estimates remain off by about 6 points. Self-improvement studies should report held-out gains with their uncertainty. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.09239 — 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 2026-10-08T10:14:05.921Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).