arXiv cs.CL· Wonjeong Jo, Gyuwan Kim, Ming Shan Hee·· 2 天前AI 评分27
改个数字就改答案：LLM 文本评分中的评分量表敏感性评估
Change the Number, Change the Answer: Assessing Rating Scale Sensitivity in LLM-Based Text Scoring
AI 导读
研究定义并系统考察了 LLM 文本评分中的"量表敏感性"，即评分量表设计导致的分数变化，并将其区分为归一化分数变化的数值敏感性与分数解读变化的语义敏感性。
正文 · 原文
Abstract:Large language models (LLMs) are increasingly used in computational social science to assign scalar scores to social and psychological constructs. However, studies use different rating scales, and whether their scores remain comparable across scale designs is understudied. To address this gap, we define and systematically investigate scale sensitivity, the variation in LLM-generated scores caused by rating scale design. We distinguish numerical sensitivity, changes in normalized scores, from semantic sensitivity, changes in score interpretation. To assess both, our experimental design crosses 2 formats (continuous vs. ordinal), 4 ranges (3, 5, 11, and 21 points), and 2 signs (signed vs. unsigned) across 3 tasks (political leaning, sentiment, and toxicity), 2 prompting strategies (zero-shot and rubric-based), and 5 LLMs. We find that changing one design element shifts normalized scores by about 0.10 on average and changes the category of roughly 20% of texts. Sensitivity is greatest for toxicity, the only unipolar construct examined. Also, rubric-based prompting does not reduce sensitivity compared with zero-shot prompting. We provide practical guidance for designing more reliable LLM-based text scoring and highlight the need for robustness checks across alternative scales.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Numerical Analysis (math.NA) |
| Cite as: | arXiv:2610.22187 [cs.CL] |
| (or arXiv:2610.22187v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22187 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Claire Wonjeong Jo [view email]
[v1]
Thu, 15 Oct 2026 16:42:15 UTC (336 KB)
来源：arXiv cs.CL · arxiv.org