语言模型与先验的能力门控池化用于事件预测
原标题：Competence-Gated Pooling of Language Models and Priors for Event Forecasting
AI 摘要
伊利诺伊大学厄巴纳-香槟分校的研究者提出了一种“能力门控池化”方法，用于在混合预测中判断语言模型相对于已有外部预测（市场、群体或统计预测）是否具有边际价值。该方法基于Brier损失推导出模型分歧何时能改善外部预测，并从已解决结果中估计领域级源权重，向全局权重收缩并进行等渗重校准。在2357个已解决二元问题和五个语言模型上，该方法将主外部基线的Brier分数从0.0771改善至0.0732，显著优于全局组合，但在强市场子集上无显著提升，且发现语言模型的言语置信度无法可靠识别其相对外部预测的优势。
正文节选
Competence-Gated Pooling of Language Models and Priors for Event Forecasting Abstract In hybrid forecasting, a language model is often one of several available signals. A system may already have a market, crowd, or statistical forecast and must decide whether the model adds useful information or should be ignored. The relevant target is therefore not standalone model accuracy, but relative competence, defined as the model’s marginal value beyond the available external forecast. Under Brier loss,