ReAct: Synergizing Reasoning and Acting in Language Models
Source: Google, Princeton University While large language models (LLMs) have demonstrated impressive capabilities across tasks in language understanding and interactive decision making, their abilities for reasoning (e.g. chain-of-thought prompting) and acting (e.g. action plan generation) have primarily been studied as separate topics. In this paper, we explore the use of LLMs to generate both reasoning […]
HEALTH
WORLD AFFAIRS
TECHNOLOGY
CREATIVE
