On the MPDD depression-detection benchmark, a compact transformer (1.06M params) outperforms XGBoost and a fine-tuned 7B LLaMA-2 model on most classification tasks.
Dynamic multimodal measurement of depression severity using deep autoencoding,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Exploring Machine Learning and Language Models for Multimodal Depression Detection
On the MPDD depression-detection benchmark, a compact transformer (1.06M params) outperforms XGBoost and a fine-tuned 7B LLaMA-2 model on most classification tasks.