LLMs develop coherent latent representations of emotions that align with valence-arousal models, show mostly linear geometry, and support uncertainty quantification in emotion tasks.
Classify this text into exactly one emotion from this list: . . . Text:{text}Emotion
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Latent Structure of Affective Representations in Large Language Models
LLMs develop coherent latent representations of emotions that align with valence-arousal models, show mostly linear geometry, and support uncertainty quantification in emotion tasks.