Metraipolis › archive › dc_172bc7bafa24
Hinted Scaling Laws — arXiv:2611.02270
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02270
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:28:08.689Z
- SHA-256 of the archived text
23b96e74bfbf46da415c8f20ce2faeb15bbf5da72d8776327de5cd51e796b035- Stable URL of this page
https://metraipolis.com/documento/dc_172bc7bafa24
How to cite
Hinted Scaling Laws — arXiv:2611.02270 — arXiv (en), https://arxiv.org/abs/2611.02270 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_172bc7bafa24 (Metraipolis archive dc_172bc7bafa24, sha256 23b96e74bfbf46da415c8f20ce2faeb15bbf5da72d8776327de5cd51e796b035)
BibTeX
@misc{dc172bc7bafa24,
title = {Hinted Scaling Laws — arXiv:2611.02270},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02270},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_172bc7bafa24; sha256 23b96e74bfbf46da415c8f20ce2faeb15bbf5da72d8776327de5cd51e796b035},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02270. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Scaling laws predict the outcome of large, expensive training runs by extrapolating trends from smaller, cheaper runs. On downstream accuracy evaluations, these predictions fail when the weaker models produced by the cheaper runs score near zero accuracy, because these scores give no trend to extrapolate to larger scales. We introduce hinted scaling laws, which use privileged information to turn a benchmark into a continuum of easier versions so that even the small models score above zero. We find that accuracy rises smoothly and predictably with the hint level, enabling new types of scaling laws that operate below the noise threshold. On the Delphi pretraining ladder, a hinted scaling law fit on models trained with at most $10^{20}$ FLOPs forecasts the GSM8K accuracy of a $10^{23}$ FLOP model (25.9%) to within 3 percentage points; a standard scaling law under-forecasts it by 13 percentage points when fit on the same models and by 12 even with models up to $10^{22}$ FLOPs in the fit. Across both this pretraining ladder and a collection of post-trained model families, hinted scaling laws reduce the median forecasting error on GSM8K, AIME and HLE from 21 percentage points to 4.3. Hinting also reduces the number of test examples needed to tell whether one model is better than another by widening the accuracy gaps between models while approximately preserving their order: a single hinted rollout per problem ranks nearly indistinguishable models as accurately as five unhinted rollouts. Our results suggest that hinting is a general primitive for improving the accuracy and cost of downstream scaling laws and other prediction tasks in model development.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2611.02270 — 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 2026-11-03T10:19:50.165Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).