arXiv cs.CL· SeungHyun Yi, Seungjun Yi, Sriram Sattiraju, Haeun Park, Joakim Nguyen, Terence Lim, Joseph Skrovan, Mehak Beri, Hitakshi Modi, Eloy Carrion, Andrew Well, Carlos M. Mery, Alexander Rasgon, Mia K Markey, Ying Ding·· 2 天前AI 评分20
社交媒体压力与情绪预测中提示词演化的有效策略识别
Identifying Effective Strategies in Prompt Evolution for Stress and Emotion Prediction
AI 导读
研究对比六种提示词演化方法在社交媒体压力与情绪数据上、两种模型规模下的表现，发现 GEPA 及其结合两个提示词指令的扩展版 GEPA+merge 在多个设置中优于初始提示词。研究还识别出五种与性能提升相关的提示策略，包括引导关注情绪证据、区分当前与过往痛苦、示例间接线索、要求情绪背景和引导标签选择。二元压力预测较早改善，而多标签情绪预测初期下降后部分恢复。
正文 · 原文
Authors:SeungHyun Yi, Seungjun Yi, Sriram Sattiraju, Haeun Park, Joakim Nguyen, Terence Lim, Joseph Skrovan, Mehak Beri, Hitakshi Modi, Eloy Carrion, Andrew Well, Carlos M. Mery, Alexander Rasgon, Mia K Markey, Ying Ding
Abstract:Early stress detection and risk prediction are crucial for timely mental health assessment and support. While large language models have recently shown promise for social media stress prediction, relying on accuracy can obscure false positives. Also, prompt revisions yield inconsistent gains across models and tasks, with unclear links to strategy combinations and instruction changes. By comparing six methods on social media stress and emotion data across two model sizes, we found that GEPA and its extension combining instructions from two prompts (GEPA+merge) improved prediction over initial prompts in several settings. We also identified five prompting strategies associated with development gains. These strategies included directing attention to emotional evidence, distinguishing current from past distress, illustrating indirect cues, requiring emotional context, and guiding label selection. Moreover, we documented four observations across tasks and evaluation settings. In a separate set of searches, binary stress prediction improved early, whereas multilabel emotion prediction initially declined and partly recovered. The other three concerned incorrect or conflicting instructions, gains on small comparison samples that did not hold on full development data, and changes in selected prompts' development and reported test rankings. These findings identify effective prompt evolution methods, strategy combinations associated with gains, and task and evaluation limits, providing guidance for applying prompt evolution methods to social media stress detection.
| Comments: | 4 figures, 2 tables. Submitted to the 2027 IEEE International Conference on Healthcare Informatics (ICHI 2027) |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.21407 [cs.CL] |
| (or arXiv:2610.21407v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.21407 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Seungjun Yi [view email]
[v1]
Thu, 15 Oct 2026 07:24:24 UTC (676 KB)
来源：arXiv cs.CL · arxiv.org