评估文档调优的Transformer表示用于个体级心理健康评估
Evaluating Document-Tuned Transformer Representations for Person-level Mental Health Assessment
AI总结：
比较基础Transformer和文档调优Transformer在个体级心理健康评估中的表现，发现文档调优模型在预测准确性和鲁棒性上显著优于基础模型。
摘要：
个体级心理评估需要聚合来自同一人的多条消息的意义，而文档级训练目标并非为此明确设计。我们在其他条件相同的情况下，对架构匹配的传统基础Transformer和文档调优Transformer（在文档级别进一步对比微调，有时称为“句子Transformer”）进行了系统的实证比较。通过比较两个纵向心理健康和心理数据集上的逐层和整体性能，我们发现文档调优模型相比基础表示表现出持续改进（Pearson r增加13.4%，p=0.015）。鲁棒性分析显示，在单词删除、同义词替换、拼写错误注入和反向翻译等扰动下，文档调优模型仍然更准确。此外，模糊语言（例如“通常”）在文档调优嵌入中更能表征结果，而丰富性（例如“很多”）在基础Transformer中更突出，这表明文档调优模型可能更好地捕捉不确定性。这些结果表明表示选择影响心理健康预测，文档调优模型通常更擅长。
英文摘要：
Person-level psychological assessment requires aggregating meaning across many messages from the same individual, a task that document-level training objectives were not explicitly designed for. We present a systematic, empirical comparison between architecturally matched traditional (a) base-transformers and (b) document-tuned-transformers (further contrastively fine-tuned at the document-level, sometimes referred to as "sentence transformers") under otherwise identical conditions. Comparing layer-wise and overall performance across two longitudinal mental health and psychological datasets, we find document-tuned models demonstrated a consistent improvement over base representations (increase in Pearson r of 13.4%, p=.015). Robustness analyses revealed document-tuned models remained more accurate under perturbations to word deletion, synonym replacement, typo injection, and back translation. Further, hedged language (e.g., `usually') was more characteristic of outcomes in document-tuned embeddings while abundance (e.g., `lot') was more characteristic of base-transformers, suggesting document-tuned models may better capture uncertainty. These results suggest representation choice impacts mental health prediction, document-tuned models often being more adept.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
