发表机构
ITMO University（ITMO大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究针对营养咨询提示方法缺乏画像分析步骤的缺陷，提出PA-CoT多阶段提示法，构建QPA基准并经对比实验证实，明确画像分析步骤可显著提升咨询的个性化与安全性。
AI 中文摘要
在健康与营养咨询领域，广泛使用的提示方法将用户画像作为非结构化块传递，未设置专门的分析步骤，这使得个性化成为关键的结构性缺陷。我们提出PA-CoT（Profile-Adaptive Chain-of-Thought，画像自适应思维链），这是一种多阶段提示方法，在生成响应前将画像解读作为明确、独立的推理步骤。为实现系统性评估，我们推出QPA（Question--Profile--Answer）基准，该基准包含200个营养咨询样本，带有结构化用户画像，并按四个标准评分。在与11种对比方法（包括CoT、Few-Shot、Role Prompting、DSPy、TextGrad、Self-Refine及零样本基线，加上PA-CoT共12种方法）的对比研究中，PA-CoT取得最佳平均得分（G-Eval 1-5量表上为4.21），且在个性化（4.71，对比最优竞品的4.39）和安全性（4.68，对比最优竞品的4.52）两项指标上领先，其95%置信区间与最优竞品无重叠，是唯一同时在这两项指标上登顶的方法。研究结果证实，与广泛使用的提示方法相比，明确的画像分析步骤是实现个性化提升的关键驱动因素。
英文摘要
In health and nutrition consulting, widely used prompting methods pass the user profile as an unstructured block without a dedicated analysis step, leaving personalization as a critical structural gap. We introduce PA-CoT (Profile-Adaptive Chain-of-Thought), a multi-stage prompting method that treats profile interpretation as an explicit, standalone reasoning step prior to response generation. To enable systematic evaluation, we introduce the QPA (Question--Profile--Answer) benchmark -- 200 nutritional consulting samples with structured user profiles scored on four criteria. In a comparative study against 11 comparison methods (CoT, Few-Shot, Role Prompting, DSPy, TextGrad, Self-Refine, and others, plus a Zero-Shot Baseline; 12 total including PA-CoT), PA-CoT achieves the best average score (4.21 on the G-Eval 1--5 scale) and leads on both Personalization (4.71 vs. 4.39) and Safety (4.68 vs. 4.52) with non-overlapping 95\% confidence intervals over the nearest competitor -- the only method to simultaneously top both criteria. The results confirm that an explicit profile-analysis step is the key driver of personalization gains over widely used prompting approaches.
CommentsAccepted at the ICML 2026 Workshop on Structured Data for Health (SD4H)
