AI Paper Atlas

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

Tracing Intricate Cues in Dialogue: Joint Graph Structure and Sentiment Dynamics for Multimodal Emotion Recognition

Jiang Li, Xiaoping Wang, Zhigang Zeng

TPAMI 2025 · Vision-language and multimodal › Cross-modal learning · approach: Multimodal sentiment analysis · 41 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 sentiment classification task (multitask L_s) to MERC training (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 71.52 -> 72.81Improved
add sentiment dynamics perception shift loss L_o (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 71.42 -> 72.81Improved
add shift loss L_o (ablation)GraphSmile · MELD (MERC)WF1: 66.59 -> 66.71Improved
add residual connection summing all layer outputs (ablation)GraphSmile · MELD (MERC)WF1: 46.24 -> 66.71Improved
add residual connection (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 61.35 -> 72.81Improved
add FC layer inside residual connection (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 70.39 -> 72.81Improved
add utterance segmentation for shift classification (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 71.69 -> 72.81Improved
add textual modality (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 54.23 -> 72.81Improved
add visual modality (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 71.23 -> 72.81Improved
add acoustic modality (ablation)GraphSmile · IEMOCAP-6 (MERC)WF1: 68.26 -> 72.81Improved
add emotion task loss L_e to sentiment task training (ablation)GraphSmile · MELD (MSAC)WF1: 73.84 -> 74.31Improved
tune hyperparameters specifically for the MSAC taskGraphSmile · IEMOCAP-4 (MSAC)ACC: 85.79 -> 87.49Improved

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