{"slug": "proper-scoring-rules-shape-llm-forecasting", "title": "Proper Scoring Rules Shape LLM Forecasting", "summary": "A new arXiv preprint (2608.28482) finds that the choice of proper scoring rule as a training objective materially shapes LLM forecasters' calibration, probability use, and error structure, even though all five rules tested share the same theoretical incentive for truthful reporting. The Brier-trained model achieved the lowest observed Brier score and highest AUC-ROC, while the log-trained model achieved the highest observed log score and lowest calibration error. The study cautions that each condition used a single seed, so some differences may reflect training stochasticity.", "body_md": "# Computer Science > Machine Learning\n\n[Submitted on 28 Aug 2026]\n\n# Title:How Proper Scoring Rules Shape LLM Forecasting\n\n[View PDF](/pdf/2608.28482)\n\n[HTML (experimental)](https://arxiv.org/html/2608.28482v1)\n\nAbstract:This paper evaluates how reward function choice shapes the performance and behavior of LLM forecasters. We compare five proper scoring rules as training objectives for binary forecasts of resolved real-world events. Although the rules share the same theoretical incentive for truthful probability reporting, the resulting models differ in calibration, probability use, and estimated profiles of bias, information, and noise, with smaller differences in aggregate accuracy and discrimination. The Brier-trained model has the lowest observed Brier score and highest AUC-ROC, while the log-trained model has the highest observed log score and lowest calibration error. Models with similar aggregate performance also reach that performance through different combinations of bias, information, and noise. Proper scoring rules therefore need not behave interchangeably as training objectives. Reward choice may shape not only how well an LLM forecasts, but how its forecasting errors are structured. Each condition uses a single seed, so some differences may reflect training stochasticity.\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/))# 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))# 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))# 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/proper-scoring-rules-shape-llm-forecasting", "canonical_source": "https://arxiv.org/abs/2608.28482", "published_at": "2026-09-02 18:59:47+00:00", "updated_at": "2026-09-02 19:24:16.458706+00:00", "lang": "en", "topics": ["machine-learning", "large-language-models", "ai-research"], "entities": ["arXiv"], "alternates": {"html": "https://wpnews.pro/news/proper-scoring-rules-shape-llm-forecasting", "markdown": "https://wpnews.pro/news/proper-scoring-rules-shape-llm-forecasting.md", "text": "https://wpnews.pro/news/proper-scoring-rules-shape-llm-forecasting.txt", "jsonld": "https://wpnews.pro/news/proper-scoring-rules-shape-llm-forecasting.jsonld"}}