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. We further apply this target through Group-Uncertainty Preference Optimization, or GUPO, an uncertainty-channel training framework that supervises verbal uncertainty expressions rather than the full response. Experiments across factual, mathematical, and multiple-choice reasoning tasks show improved uncertainty ranking, lower calibration error, and reduced overconfidence.

翻译：大语言模型日益通过自然语言陈述来表达不确定性，然而这类表达往往未能反映模型的采样行为。我们将言语不确定性对齐视为一个分布校准问题：提示的适当不确定性目标应从重复模型输出中估计，而非孤立响应。然而仅依靠组别展开并不足够，因为由此产生的目标必须提供有效的训练信号。现有目标仅部分满足这一要求。我们提出SAGE（语义答案引导熵），这是一种组级不确定性目标，通过在采样响应上构建答案条件化的不确定性几何结构。SAGE在保持平滑且尺度不变的校准信号的同时，保留了类别型、数值型和符号型答案的区分特征。我们进一步通过组别不确定性偏好优化（GUPO）框架应用该目标，该框架作为不确定性通道训练框架监督言语不确定性表达而非完整响应。在事实型、数学型及多项选择推理任务上的实验表明，该方法改善了不确定性排序、降低了校准误差并减少了过度自信现象。