无 Transformer 胜过六个协变量：基于童年作文的抑郁症状长期预测

中文摘要 AI 辅助

自然语言处理（NLP）模型能够检测在症状测量时间附近撰写的文本中的抑郁相关语言，但预训练 Transformer 能否从十二年前撰写的文本中预测抑郁症状，在很大程度上尚未得到测试。在英国出生队列研究——国家儿童发展研究中，我们利用同一批人在 11 岁时撰写的作文，预测其在 23 岁时可能存在的抑郁症状。我们的基线模型，即基于六个童年协变量的逻辑回归，优于所有仅查看作文的文本模型：七个微调 Transformer、一个词袋模型、冻结嵌入以及四个零样本大型语言模型。其受试者工作特征曲线下面积（AUC-ROC）为 0.737，而主要种子上最佳 Transformer 的 AUC-ROC 为 0.670，且没有任何附加文本评分能显著提高基线的 AUC-ROC。经 Bonferroni 校正后，五个领域预训练 Transformer 均未显著优于其通用领域对照组。对于长期预测，基线模型仍是需要被超越的模型。

英文摘要

Natural language processing (NLP) models can detect depression-related language in text written near the time symptoms are measured, but whether pretrained transformers can predict depressive symptoms from text written twelve years earlier is largely untested. In the National Child Development Study, a British birth cohort, we predict probable depressive symptoms at age 23 from essays the same people wrote at age 11. Our baseline, a logistic regression on six childhood covariates, outperforms every text model that sees only the essay: seven fine-tuned transformers, a bag-of-words model, frozen embeddings and four zero-shot large language models. Its area under the receiver operating characteristic curve (AUC-ROC) is 0.737 against 0.670 for the best transformer on the primary seed, and no added text score detectably raises the baseline's AUC-ROC. None of the five domain-pretrained transformers detectably beats its general-domain control after Bonferroni correction. For long-horizon prediction, the baseline remains the model to beat.

发表机构

• University of British Columbia（不列颠哥伦比亚大学）
• Vector Institute（向量研究所）

机构由 AI 辅助整理，请以论文原文为准。