However, many questions remained about whether these models could perform human-like visual reasoning." The main objective of the recent study by Buschoff, Akata and their colleagues was to assess ...