做社会科学量化分析的团队终于有了校准LLM输出的实操方案——软标签蒸馏能显著降低置信度偏差，建议做文本编码和实证研究的点开看看具体方法。
大型语言模型（LLM）越来越多地被用于社会科学研究，将非结构化文本转换为可进入实证设计的变量。但研究发现，LLM的置信度与真实正确率严重不匹配，导致基于置信度过滤会扭曲下游回归估计。研究对GPT-5-mini、DeepSeek-V3.2等14个社会科学构念进行审计，发现所有模型都存在校准偏差。作为解决方案，提出软标签蒸馏管道，将LLM得分和置信度转化为软目标分布，训练小型判别分类器，平均降低ECE 43.2%和Brier 34.0%。研究呼吁将校准视为测量有效性的组成部分，而非可选的后期处理。
Assessing and Mitigating Miscalibration in LLM-Based Social Science Measurement
Large language models (LLMs) are increasingly used in social science as scalable measurement tools for converting unstructured text into variables that can enter standard empirical designs. Measurement validity demands more than high average accuracy, which requires well calibrated confidence that faithfully reflects the empirical probability of each measurement being correct. This paper studies the model miscalibration in LLM-based social science measurement. We begin with a case study on FOMC and show that confidence based filtering can change downstream regression estimates when LLM confidence is miscalibrated. We then audit calibration across 14 social science constructs covering both proprietary models, including GPT-5-mini, DeepSeek-V3.2, and open source models. Across tasks and model families, reported confidence is poorly aligned with tolerance-based correctness. As a simple mitigation, we propose a soft label distillation pipeline for calibrating Bert with LLM. The method converts an LLM score and its verbalized confidence into a soft target distribution, then trains a smaller discriminative classifier on encoder models for these targets. Averaged across datasets, this approach reduces ECE by 43.2\% and Brier by 34.0\%. These results suggest that LLM-based social science pipelines should treat calibration as part of measurement validity, rather than as an optional post-processing concern.