Metraipolis › archive › dc_4b2fcb2460ee

The Free-Recipe Limit: Every Recipe Effect Measures Which Premise of an Idealised Learner Broke — arXiv:2609.26160

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.26160
License (as archived)
CC0-1.0
Archived
2026-10-08T13:45:36.790Z
SHA-256 of the archived text
92b2db481560309eee9a2281cc3fc9fe7e726c5cda6b7be7e8b7ae16c1b0a3e9
Stable URL of this page
https://metraipolis.com/documento/dc_4b2fcb2460ee

How to cite

The Free-Recipe Limit: Every Recipe Effect Measures Which Premise of an Idealised Learner Broke — arXiv:2609.26160 — arXiv (en), https://arxiv.org/abs/2609.26160 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4b2fcb2460ee (Metraipolis archive dc_4b2fcb2460ee, sha256 92b2db481560309eee9a2281cc3fc9fe7e726c5cda6b7be7e8b7ae16c1b0a3e9)
BibTeX
@misc{dc4b2fcb2460ee,
  title = {The Free-Recipe Limit: Every Recipe Effect Measures Which Premise of an Idealised Learner Broke — arXiv:2609.26160},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.26160},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4b2fcb2460ee; sha256 92b2db481560309eee9a2281cc3fc9fe7e726c5cda6b7be7e8b7ae16c1b0a3e9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.26160. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Fix a corpus and send recipe search to infinity: try every order of the skills, every arrangement from blocked to interleaved, every composition, and keep the best. Two quantities decide what that search was worth: the diameter of the reachable set it explores, and the resolution at which anyone can tell two endpoints apart. Where the diameter falls below the resolution, no amount of search converts into a decision, and the signature is not an absence of winners but winners that do not survive re-running. We measure this recipe-search wall with 761 fine-tuning runs on 12 base models (0.5B-14B, three pretraining families) over competition-mathematics skills: base checkpoints, supervised fine-tuning under AdamW, exact-match scoring at k=4. Within one coherent domain at fixed volume the three classical freedoms average 0.010-0.021 against a 0.019 floor, and the largest contrast, 0.0619, clears a three-seed resolution and then reads +0.010 and -0.015 on two reruns. The departure with a systematic answer is coherence: halving one pooled corpus and letting the halves write answers under incompatible but equally correct conventions moves arrangement from capability to allocation between conventions, by two orders of magnitude over a same-convention control, and writing the convention into the input switches the phenomenon off. The switch replicates on a second pretraining family and survives an independent re-execution of its own protocol, with a re-execution spread (0.087) smaller than the resolution a search-selected order cell carries (0.144). Order itself is a transient whose sign crosses zero three times inside a single run. Volume, the one lever nobody calls a recipe, is the one that reliably pays. A public scorecard grades all 26 pre-registered claims: 18 supported, 5 failed, 2 untested, 1 mixed.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-23 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.26160 — 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 2026-10-08T13:18:38.153Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-23?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).