LLMs’ “simulated reasoning” abilities are a “brittle mirage,” researchers find
Credit: Zhao et al The researchers used test cases that fall outside of the LLM training data in task type, format, and length. Credit: Zhao et al The researchers used test cases that fall outside of the LLM training data in task type, format, and length. These simplified models were then tested using a variety of tasks, some of which precisely or closely matched the function patterns in the training data and others that required function compositions that were either partially or fully "out of