AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Voices, Faces, and Feelings: Multi-modal Emotion-Cognition Captioning for Mental Health Understanding

Zhiyuan Zhou, Yanrong Guo, Shijie Hao

AAAI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal sentiment analysis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ECMC emotion-cognition profile as input to an LLM for depression detectionQwen3-8B · Depression detectionACC (%): 50.46 -> 70.64Improved
add ECMC profile for anxiety detectionLlama-3-8B · Anxiety detectionACC (%): 38.54 -> 57.29Improved
add ECMC profile for anxiety detectionGLM4-9B · Anxiety detectionACC (%): 43.75 -> 44.95No clear effect
use profiles from weaker captioners (e.g. MindChat) instead of noneLLM detectors · Depression/anxiety detectionF1: 1.01 decrease in avg F1 gain (MindChat)Hurt
combine audio, video and text modalities rather than single modality (ablation)ECMC · MMDA captioningF BERTImproved
add both emotion and cognition contrastive learning branches (ablation)ECMC · MMDA captioningF BERT: full model 34.09Improved
use multimodal ECMC vs text-only mental health LLMs for caption generationECMC · MMDA emotion captioningBLEU-4: 2.33 -> 8.28Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy