RPCL is a training-only method that raises mean Pair F1 by 2.58-2.83 points on three MECPE datasets by enforcing larger gold-negative confidence gaps and prediction stability under context corruption.
Multimodal emotion recognition in conversation with mutual information maximization and contrastive loss
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Learning Robust Pair Confidence for Multimodal Emotion-Cause Pair Extraction
RPCL is a training-only method that raises mean Pair F1 by 2.58-2.83 points on three MECPE datasets by enforcing larger gold-negative confidence gaps and prediction stability under context corruption.