arXiv cs.AI· Zhaomin Wu, Jiayi Li, Bingsheng He·· 11 小时前AI 评分29
EmoTrack：面向文本抑郁严重程度评估的临床语义建模
EmoTrack: Clinical-Semantic Modeling for Text-Based Depression Severity Estimation
AI 导读
EmoTrack 联合编码临床特征 token 与冻结的来访者对话轮嵌入，并用 PHQ-8 条目查询估计抑郁严重程度总分，可选门控记忆引入前次会话证据。在 DAIC-WOZ 上，其平均绝对误差相对最强基线降低 12.5%，聚合误差降低约 8.6%。团队同时发布带会话级 PHQ-8 监督的合成咨询数据集 CounselArc，用于多会话场景的受控评估。
正文 · 原文
Abstract:Text-based counseling is widely adopted across online mental health platforms, where automatically estimating depressive symptom severity from transcripts is vital for client safety and clinical monitoring. We therefore study prediction of the standard eight-item Patient Health Questionnaire (PHQ-8) total score directly from counseling sessions. Existing clinical-based methods obtain structured assessments through large language model inference but summarize away individual client turns. Training-based methods preserve turn semantics but must learn clinical structure from limited labels. Integrating these representations requires preserving clinical-feature identity and selecting evidence relevant to each symptom. We propose EmoTrack, which jointly encodes clinical-feature tokens and frozen client-turn embeddings, then uses PHQ-8 item queries to estimate total severity. Optional gated memory incorporates evidence from the preceding session. To support controlled evaluation in the multi-session setting, we introduce CounselArc, a synthetic counseling dataset with session-level PHQ-8 supervision for predicting current-session severity with optional prior-session context. Across real DAIC-WOZ interviews, transfer to E-DAIC, and three longitudinal datasets, EmoTrack achieves the lowest normalized aggregate error. It reduces DAIC-WOZ mean absolute error by 12.5% and aggregate error by approximately 8.6%, each relative to the strongest baseline for that measure.
| Comments: | 36 pages. Related LongCounsel-8 benchmark: arXiv:2609.03507 |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.22286 [cs.LG] |
| (or arXiv:2605.22286v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.22286 arXiv-issued DOI via DataCite |
Submission history
From: Jiayi Li [view email]
[v1]
Thu, 21 May 2026 10:37:45 UTC (576 KB)
[v2]
Tue, 8 Sep 2026 13:05:25 UTC (534 KB)
[v3]
Wed, 4 Nov 2026 14:41:55 UTC (302 KB)
来源：arXiv cs.AI · arxiv.org