SAGE: Answer-Conditioned Uncertainty Targets for Verbal Uncertainty Alignment 文章

ArXiv CS.CL2026-06-11NEWSen作者: Kaiwen Shi, Zheyuan Zhang, Yanfang Ye

详细信息

来源站点
ArXiv CS.CL
作者
Kaiwen Shi, Zheyuan Zhang, Yanfang Ye
文章类型
NEWS
语言
en
发布日期
2026-06-11

摘要

arXiv:2606.11512v1 Announce Type: new Abstract: Large language models increasingly express uncertainty through natural-language statements, yet these expressions often fail to reflect the model's sampled behavior. We study verbal uncertainty alignment as a distributional calibration problem: the appropriate uncertainty target for a prompt should be estimated from repeated model outputs rather than from an isolated response. However, group rollouts alone are insufficient, since the resulting target must provide a useful training signal. Existing targets only partially satisfy this requirement. We propose SAGE, Semantic-Answer Guided Entropy, a group-level uncertainty target that constructs an answer-conditioned uncertainty geometry over sampled responses. SAGE preserves categorical, numeric, and symbolic answer distinctions while maintaining a smooth and scale-preserving calibration signal.

相关事件

暂无数据

相关公司

暂无数据

相关人物

暂无数据

相关产品

暂无数据