Chehre: An Emoji-Prompted Dataset to Explore Perceptual Flexibility in Video Language Models
Chehre, an emoji-prompted video dataset with a wide range of dynamic facial expressions for exploring perceptual variation, finds that persona prompting can act as a controllable way to shift model perception while helping models better capture the variation observed across human annotators.
Bita Azari, Zoe Stanley, Avneet Batra et al.
· 0 citations