LLM-BABYBENCH: Understanding and Evaluating Grounded Planning and Reasoning in LLMs

Omar Choukrani, Idriss Malek, Daniil Orel +4

Assessing the capacity of Large Language Models (LLMs) to plan and reason within the constraints of interactive environments is crucial for developing capable AI agents. We introdu…