BehavDep Framework for Interpretable Depression Assessment
October 9, 2026
BehavDep is a sparse factor-based framework for multimodal depression assessment that decomposes behavioral representations into interpretable latent factors. It uses weak supervision to bridge the gap between video-level data and user-level clinical annotations.
HOW THIS AFFECTS YOU
●
researcherYou can utilize sparse representations to improve the transparency of multimodal models.
●
healthThis provides more interpretable behavioral insights for digital mental health tools.