In this competition, you’ll develop AI systems to efficiently learn new skills and solve open-ended problems, rather than depend exclusively on AI systems trained with extensive datasets. The top submissions will show improvement toward human reasoning benchmarks.