Multimodal LLMs capture the coarse category-level structure of human emotional responses to videos, but not fine-grained item-level emotion structure.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Correspondence of high-dimensional emotion structures elicited by video clips between humans and Multimodal LLMs
Multimodal LLMs capture the coarse category-level structure of human emotional responses to videos, but not fine-grained item-level emotion structure.