Introducing NGPQA, a new benchmark on which o3 gets only 12% whereas a toddler, a dog, and even an *ant* get perfect scores
This benchmark shows what Yann LeCun has been saying all along--that these LLMs aren't even at dog intelligence, much less the golden spark of human intelligence!
NGPQA (which stands for Negated GPQA) is a simple extension of the popular GPQA benchmark; it simply takes GPQA scores and runs them through the function f(x)=(100-x)