Visual Riddles: a Commonsense and World Knowledge Challenge for Large Vision and Language Models | Read Paper on Bytez