activity
20172026
most citedVLMbench: A Compositional Benchmark for Vision-and-Language Manipulation

16 citations · 57 across the 46 of their papers we have counts for

collaborators
Showing 2019Show all

4 papers · 1 filter