Capturing Failures of Large Language Models via Human Cognitive Biases

Open in new window