Benchmarks are way too easy to game. There's no shortage of models that "beat GPT-4" according to some benchmark or another, that are obviously nowhere even clo

Benchmarks are way too easy to game. There's no shortage of models that "beat GPT-4" according to some benchmark or another, that are obviously nowhere even close when you try them on novel tasks.