Reasoning Capabilities of Large Language Models on Dynamic Tasks

Annie Wong, Thomas Bäck, Aske Plaat +2

Large language models excel on static benchmarks, but their ability as self-learning agents in dynamic environments remains unclear. We evaluate three prompting strategies: self-re…