1 paper
Asmar Nadeem, Faegheh Sardari, Robert Dawes +3
Existing video captioning benchmarks and models lack causal-temporal narrative, which is sequences of events linked through cause and effect, unfolding over time and driven by char…