arXiv Computation and Language September 21, 2026 By Qingkun Deng, Saturnino Luz, Sofia de la Fuente Garcia

Hierarchical attention interpretation: an interpretable speech-level transformer for bi-modal depression detection

Read the original on arXiv Computation and Language →

The paper presents a bi‑modal speech‑level transformer that eliminates segment‑level labeling and introduces a hierarchical attention interpretation method. By using gradient‑weighted attention maps from all attention layers, the model provides both speech‑level and sentence‑level explanations of depression detection. Experimental results show the transformer outperforms a segment‑level model (p=0.854 vs. 0.732, r=0.947 vs. 0.808, F1=0.897 vs. 0.768).

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv AI

Jul 28

Multimodal Domain Generalization for Depression Detection: An Attention-Based BiLSTM Network with Domain-Adversarial Training

arXiv:2607. 22794v1 Announce Type: cross Abstract: Automatic depression detection with deep learning has shown promise but often suffers from limited generalization due to domain shift arising from inter-speaker variability.

By Ali Tabaraei, Federico Simonetta, Stavros Ntalampiras

multimodal benchmarks safety

arXiv AI

Jun 11

MA-DLE: Speech-based Automatic Depression Level Estimation via Memory Augmentation

arXiv:2606. 11197v1 Announce Type: cross Abstract: Speech-based automatic estimation of depression levels is essential for enabling early detection and timely intervention, particularly in resource-constrained mental health settings.

By Xuzhi Wang, Xinran Wu, Ziping Zhao, Jianhua Tao, Bj\"orn W. Schuller

benchmarks

arXiv AI

Sep 3

Interpretable Symptom Vectors for Depression in a Large Language Model

The study investigates how a large language model, Gemma-3-27B-PT, internally represents depressive symptoms. By applying mechanistic interpretability methods to the model’s residual stream, researchers found that symptom groups are geometrically distinct at layer 21, and that projected symptom vectors align with clinician-annotated rankings across mood, somatic, and suicidality dimensions. Additionally, a single depression vector at this layer can differentiate depressive from non-depressive text with an AUC of 0.789, suggesting a potential emotional valence gate for symptom projection.

By Fangyi Zhu, Ajay Subramanian, Allison Constant, Camille Wang, Ravish Gupta, Corey J. Keller

llms safety

arXiv Computation and Language

Aug 21

Explainable Multimodal Depression Recognition in Clinical Interviews via PHQ-Aligned Symptom Summarization

arXiv:2501. 16106v2 Announce Type: replace Abstract: Recent advances in multimodal depression recognition for clinical interviews (MDRC) have demonstrated the potential of AI systems by integrating textual, acoustic, and facial cues.

By Wenjie Zheng, Qiming Xie, Jianfei Yu, Yang Wang, Lei Cao, Fei Wang, Shijin Wang, Rui Xia, Chengqing Zong

nlp multimodal safety

arXiv AI

Sep 10

EmoTrack: Clinical-Semantic Modeling for Text-Based Depression Severity Estimation

arXiv:2605.22286v2 Announce Type: replace-cross Abstract: Text-based counseling provides a valuable source of information for assessing depression severity. We study prediction of the total score on...

By Zhaomin Wu, Jiayi Li, Bingsheng He

llms rag benchmarks

arXiv Machine Learning

Jun 25

Expresso-AI: Explainable Video-Based Deep Learning Models for Depression Diagnosis

arXiv:2606. 25606v1 Announce Type: cross Abstract: Given the widespread prevalence of depression and its consequential impact on individuals and society, it is crucial to obtain objective measures for early diagnosis and intervention.

By Felipe Moreno, Sharifa Alghowinem, Hae Won Park, Cynthia Breazeal

fine-tuning benchmarks safety

One email a morning, machine-written

Related stories

Multimodal Domain Generalization for Depression Detection: An Attention-Based BiLSTM Network with Domain-Adversarial Training

MA-DLE: Speech-based Automatic Depression Level Estimation via Memory Augmentation

Interpretable Symptom Vectors for Depression in a Large Language Model

Explainable Multimodal Depression Recognition in Clinical Interviews via PHQ-Aligned Symptom Summarization

EmoTrack: Clinical-Semantic Modeling for Text-Based Depression Severity Estimation

Expresso-AI: Explainable Video-Based Deep Learning Models for Depression Diagnosis