English(EN) How Proper Scoring Rules Shape LLM Forecasting

论文：奖励函数选择显著影响 LLM 预测行为

作者 PulseAugur 编辑部 · [1 个来源] · 2026-08-31 04:00

一篇新论文探讨了不同的奖励函数（即 proper scoring rules）如何影响大型语言模型的预测能力和行为。尽管这些规则理论上鼓励诚实的概率报告，但研究发现，使用不同规则训练的模型在校准、概率使用以及偏差、信息和噪声的估计画像方面表现出差异。Brier 训练的模型在 Brier 分数和 AUC-ROC 方面表现最佳，而 log 训练的模型在 log 分数和校准误差方面表现出色，这表明奖励函数的选择不仅显著影响预测准确性，还影响预测错误的结构。 AI

影响这项研究强调，LLM 训练目标的选择会显著改变其预测能力和错误结构，这表明在开发可靠的 AI 预测器时需要仔细考虑奖励函数。

排序理由该项目是一篇学术论文，详细介绍了对 LLM 行为的研究。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 14 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该项目是一篇学术论文，详细介绍了对 LLM 行为的研究。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Benjamin Turtel, Paul Wilczewski, Kris Skotheim, Ville A. Satop\"a\"a, Philip E. Tetlock · 2026-08-31 04:00

Proper Scoring Rules 如何塑造 LLM 预测

arXiv:2608.28482v1 Announce Type: new Abstract: 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 rul…

本文如何被排名

报道来源 [1]

Proper Scoring Rules 如何塑造 LLM 预测

相关实体

相关话题