LLMs given only valence and arousal values classify facial expressions poorly but can generate free-text emotion descriptions that align with human annotations under Word2Vec and BERT similarity, though not under a generic Transformer embedding.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Beyond Vision: How Large Language Models Interpret Facial Expressions from Valence-Arousal Values
LLMs given only valence and arousal values classify facial expressions poorly but can generate free-text emotion descriptions that align with human annotations under Word2Vec and BERT similarity, though not under a generic Transformer embedding.