{"slug": "hinted-scaling-laws", "title": "Hinted Scaling Laws", "summary": "A November 1, 2026 arXiv paper by Suze Van Adrichem introduces \"hinted scaling laws,\" which use privileged information to create a continuum of easier benchmark versions so small models score above zero and trends become extrapolable. On the Delphi pretraining ladder, a hinted scaling law fit on models trained with at most 10^20 FLOPs forecast the GSM8K accuracy of a 10^23 FLOP model (25.9%) to within 3 percentage points, while a standard scaling law under-forecast it by 13 percentage points on the same models and by 12 even with models up to 10^22 FLOPs in the fit. Across the pretraining ladder and post-trained model families, hinted scaling laws cut median forecasting error on GSM8K, AIME and HLE from 21 percentage points to 4.3, and a single hinted rollout per problem ranks near-identical models as accurately as five unhinted rollouts.", "body_md": "# Computer Science > Machine Learning\n\n  [Submitted on 1 Nov 2026]\n\n# Title:Hinted Scaling Laws\n\n[View PDF](https://arxiv.org/pdf/2611.02270)\n\n[HTML (experimental)](https://arxiv.org/html/2611.02270v1)\n\nAbstract: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.\n    \n\n## Submission history\n\nFrom: Suze Van Adrichem [\n[view email](https://arxiv.org/show-email/5d843d56/2611.02270)]\n\n**[v1]** Sun, 1 Nov 2026 22:50:43 UTC (1,120 KB)\n\n### Additional Features\n\n### Current browse context:\n\ncs.LG\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\nIArxiv Recommender\n\n*(*[What is IArxiv?](https://iarxiv.org/about))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/hinted-scaling-laws", "canonical_source": "https://arxiv.org/abs/2611.02270", "published_at": "2026-11-04 00:00:11+00:00", "updated_at": "2026-11-04 00:20:45.684483+00:00", "lang": "en", "topics": ["machine-learning", "ai-research", "large-language-models", "artificial-intelligence"], "entities": ["Suze Van Adrichem", "arXiv", "Delphi", "GSM8K", "AIME", "HLE"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/hinted-scaling-laws", "markdown": "https://wpnews.pro/news/hinted-scaling-laws.md", "text": "https://wpnews.pro/news/hinted-scaling-laws.txt", "jsonld": "https://wpnews.pro/news/hinted-scaling-laws.jsonld"}}