Using Rank-N-Contrast loss on valence-arousal rankings instead of symmetric cross-entropy improves ordinal consistency and cross-modal alignment for emotional speech and text.
Don’t classify ratings of affect; rank them!
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
EmotionRankCLAP: Bridging Natural Language Speaking Styles and Ordinal Speech Emotion via Rank-N-Contrast
Using Rank-N-Contrast loss on valence-arousal rankings instead of symmetric cross-entropy improves ordinal consistency and cross-modal alignment for emotional speech and text.