Papers › Tracing Intricate Cues in Dialogue: Joint Graph Structure and Sentiment Dynamics for...

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

31 Jul 2024arXiv:2407.21536archive 2025-07-28

Jiang Li, XiaoPing Wang, Zhigang Zeng

Multimodal emotion recognition in conversation (MERC) has garnered substantial research attention recently. Existing MERC methods face several challenges: (1) they fail to fully harness direct inter-modal cues, possibly leading to less-than-thorough cross-modal modeling; (2) they concurrently extract information from the same and different modalities at each network layer, potentially triggering conflicts from the fusion of multi-source data; (3) they lack the agility required to detect dynamic sentimental changes, perhaps resulting in inaccurate classification of utterances with abrupt sentiment shifts. To address these issues, a novel approach named GraphSmile is proposed for tracking intricate emotional cues in multimodal dialogues. GraphSmile comprises two key components, i.e., GSF and SDP modules. GSF ingeniously leverages graph structures to alternately assimilate inter-modal and intra-modal emotional dependencies layer by layer, adequately capturing cross-modal cues while effectively circumventing fusion conflicts. SDP is an auxiliary task to explicitly delineate the sentiment dynamics between utterances, promoting the model's ability to distinguish sentimental discrepancies. Furthermore, GraphSmile is effortlessly applied to multimodal sentiment analysis in conversation (MSAC), forging a unified multimodal affective model capable of executing MERC and MSAC tasks. Empirical results on multiple benchmarks demonstrate that GraphSmile can handle complex emotional and sentimental patterns, significantly outperforming baseline models.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

lijfrank-open/GraphSmile officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Emotion RecognitionEmotion Recognition in ConversationMultimodal Emotion RecognitionMultimodal Sentiment AnalysisSentiment Analysis

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Emotion Recognition in Conversation CMU-MOSEI-Sentiment GraphSmile Accuracy 46.82 #1 of 7 Archive leaderboard report
Emotion Recognition in Conversation CMU-MOSEI-Sentiment GraphSmile Weighted F1 44.93 #1 of 7 Archive leaderboard report
Emotion Recognition in Conversation CMU-MOSEI-Sentiment-3 GraphSmile Accuracy 67.73 #1 of 1 Archive leaderboard report
Emotion Recognition in Conversation CMU-MOSEI-Sentiment-3 GraphSmile Weighted F1 66.73 #1 of 1 Archive leaderboard report
Emotion Recognition in Conversation IEMOCAP GraphSmile Accuracy 72.77 #2 of 59 Archive leaderboard report
Emotion Recognition in Conversation IEMOCAP GraphSmile Weighted-F1 72.81 #2 of 59 Archive leaderboard report
Emotion Recognition in Conversation IEMOCAP-4 GraphSmile Accuracy 86.53 #1 of 8 Archive leaderboard report
Emotion Recognition in Conversation IEMOCAP-4 GraphSmile Weighted F1 86.52 #1 of 8 Archive leaderboard report
Emotion Recognition in Conversation MELD GraphSmile Accuracy 67.70 #16 of 68 Archive leaderboard report
Emotion Recognition in Conversation MELD GraphSmile Weighted-F1 66.71 #16 of 68 Archive leaderboard report
Emotion Recognition in Conversation MELD-Sentiment GraphSmile Accuracy 74.44 #1 of 1 Archive leaderboard report
Emotion Recognition in Conversation MELD-Sentiment GraphSmile Weighted F1 74.31 #1 of 1 Archive leaderboard report
Multimodal Emotion Recognition CMU-MOSEI-Sentiment GraphSmile Accuracy 46.82 #1 of 1 Archive leaderboard report
Multimodal Emotion Recognition CMU-MOSEI-Sentiment GraphSmile Weighted F1 44.93 #1 of 1 Archive leaderboard report
Multimodal Emotion Recognition CMU-MOSEI-Sentiment-3 GraphSmile Accuracy 67.73 #1 of 1 Archive leaderboard report
Multimodal Emotion Recognition CMU-MOSEI-Sentiment-3 GraphSmile Weighted F1 66.73 #1 of 1 Archive leaderboard report
Multimodal Emotion Recognition IEMOCAP GraphSmile Accuracy 72.77 #1 of 2 Archive leaderboard report
Multimodal Emotion Recognition IEMOCAP GraphSmile Weighted F1 72.81 #1 of 2 Archive leaderboard report
Multimodal Emotion Recognition IEMOCAP-4 GraphSmile Accuracy 86.53 #1 of 11 Archive leaderboard report
Multimodal Emotion Recognition IEMOCAP-4 GraphSmile Weighted F1 86.52 #1 of 11 Archive leaderboard report
Multimodal Emotion Recognition MELD GraphSmile Accuracy 67.70 #1 of 3 Archive leaderboard report
Multimodal Emotion Recognition MELD GraphSmile Weighted F1 66.71 #1 of 3 Archive leaderboard report
Multimodal Emotion Recognition MELD-Sentiment GraphSmile Accuracy 74.44 #1 of 1 Archive leaderboard report
Multimodal Emotion Recognition MELD-Sentiment GraphSmile Weighted F1 74.31 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AttentionSoftmax

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections