EnvGen: Generating and Adapting Environments via LLMs for Training Embodied Agents

arXiv:2403.12014v1 Announce Type: cross
Abstract: Recent SOTA approaches for embodied learning via interaction directly employ large language models (LLMs) as agents to determine the next steps in an environment. Due to their world knowledge and reasoning capabilities, LLM agents achieve stronger performance than previous smaller agents based on reinforcement learning (RL); however, frequently calling LLMs is slow and expensive. Instead of directly employing LLMs as agents, can we use LLMs' reasoning capabilities to adaptively create training environments to help smaller …

abstract agents arxiv capabilities cs.ai cs.cl cs.lg embodied environment environments knowledge language language models large language large language models llm llms next performance reasoning reinforcement sota training type via world