New AtmosERC framework models dialogue-level affective atmosphere for emotion recognition

Researchers have developed AtmosERC, a novel graph-based framework for emotion recognition in conversations. This system models dialogues as conversational graphs to capture a latent "affective atmosphere" that influences emotional patterns. AtmosERC uses a relation-aware graph extractor to generate dialogue-level and speaker-conditioned affective priors, which then guide sequential emotion prediction. The framework has demonstrated improvements in predicting emotions in conversations, enhances LLM-based ERC when used as a plug-in cue, and provides more stable predictions even with local emotional deviations. AI

IMPACT Introduces a novel approach to modeling conversational dynamics for improved emotion recognition, potentially enhancing human-AI interaction.

RANK_REASON Academic paper detailing a new framework for emotion recognition in conversations. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• arXiv
• AtmosERC
• emotion recognition in conversation
• Republican Left of Catalonia

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →