Tag
EviDial introduces a benchmark that separates label agreement from evidence-grounded assessment in depression dialogue evaluation, revealing that 82.8% of participant-symptom cells lack sufficient textual evidence and that 69.2–96.9% of seemingly correct LLM predictions are actually unsupported by the dialogue.
MERID 是一个多模态抑郁症分析框架，通过经验驱动的递归自我改进智能体自动探索和优化检测流水线，在多个抑郁症基准上取得领先结果。
The paper introduces a transparent framework that maps acoustic speech features to DSM-5 depression indicators for interpretable detection, running locally on commodity hardware to preserve privacy.
HiMA-MDD introduces a hierarchical multi-agent system for interpreting multimodal clinical interviews to detect depression, achieving state-of-the-art performance on the E-DAIC dataset.
A novel patient-independent multimodal framework for automatic depression detection integrates BiLSTM with intra- and cross-modal attention and domain-adversarial training to improve generalization across speakers. It achieves state-of-the-art accuracy of 93.2% on the Androids-Corpus dataset.
This paper investigates the use of large language models (LLMs) and supervised classifiers for depression detection from social media text, proposing a prompt-based embedding method that enhances interpretability. Experiments on multiple datasets show that zero-shot LLMs perform well for binary classification but struggle with fine-grained severity, while supervised models on LLM summary embeddings achieve more consistent performance across multi-class and ordinal tasks.
This paper proposes PTTSD, a probabilistic framework for depression severity detection from clinical interview transcripts that models uncertainty and provides temporal interpretability, achieving competitive performance on benchmark datasets.
Proposes a fine-grained multimodal framework with a Binary Advantage-weighting Ranking Loss for automatic depression detection, achieving state-of-the-art results on D-vlog and LMVD datasets.
This paper investigates the use of conversational temporal dynamics (turn-pair timing) as a lightweight modality for automatic depression detection from dyadic clinical interviews, showing that a compact 24-dimensional timing module achieves strong performance and complements standard acoustic and semantic features when fused.
This paper introduces CLeaD, a supervised contrastive alignment framework for cross-lingual depression detection from speech using WavLM embeddings. It reveals that previous results were inflated due to speaker identity leakage and achieves modest improvements on Mandarin speakers.
This paper introduces DEPOOL, a controlled benchmark evaluating six temporal aggregation architectures across six frozen speech backbones for depression detection in dyadic interactions, finding that many configurations collapse into single-class predictions and that robustness should be a key criterion.
This paper introduces Score-Guided Classification (SGC), a framework that models pathological priors using an unsupervised generative network for EEG-based depression detection, avoiding synthetic data augmentation and improving classification accuracy.
This paper presents a hybrid model combining DistilBERT embeddings with Holographic Reduced Representation vectors encoding cognitive-linguistic features (first-person pronouns, absolutist words, negative emotion ratios) to detect depression in Reddit posts, achieving a macro F1 of 0.94 and demonstrating that theory-driven features complement contextual embeddings for explainable mental health NLP.
This paper compares several post-hoc explainability methods applied to an InceptionTime model for EEG-based depression detection, finding partial convergence among methods while highlighting methodological variability and limitations.
This paper audits benchmark evaluation in clinical-interview depression detection through four complementary probes across five datasets, finding that standard evaluation protocols may overestimate model performance and that leaderboard rankings lack stability.
Proposes an agentic framework using LangChain agents for population-scale mental health screening, focusing on depression detection from clinical transcripts. The framework incrementally locks validated stages and uses proxy-guided evaluation to ensure trustworthiness and adaptability.
Researchers present a zero-shot LLM system that assesses depression risk from Reddit posts, achieving competitive F1 scores and demonstrating scalable mental-health monitoring.