AcT2I: Evaluating and Improving Action Depiction in Text-to-Image Models.

Vatsal Malaviya, Agneet Chatterjee, Maitreya Patel, Yezhou Yang, Chitta Baral

Browse the full EMNLP paper archive.