Benchmarking Vision, Language, and Action Models on Robotic Learning Tasks multinet.ai 11 points by hsikka a day ago