A Survey of Reasoning with Foundation Models

TL;DR

This survey reviews recent progress in reasoning capabilities of foundation models like LLMs, highlighting advancements, challenges, and future directions to enhance their role in developing Artificial General Intelligence.

Contribution

It provides a comprehensive overview of foundation models' reasoning abilities, recent advancements, and potential future research directions in the field.

Findings

Foundation models exhibit significant reasoning abilities.

Recent advancements have improved reasoning performance.

Future research directions include multimodal learning and autonomous agents.

Abstract

Reasoning, a crucial ability for complex problem-solving, plays a pivotal role in various real-world settings such as negotiation, medical diagnosis, and criminal investigation. It serves as a fundamental methodology in the field of Artificial General Intelligence (AGI). With the ongoing development of foundation models, e.g., Large Language Models (LLMs), there is a growing interest in exploring their abilities in reasoning tasks. In this paper, we introduce seminal foundation models proposed or adaptable for reasoning, highlighting the latest advancements in various reasoning tasks, methods, and benchmarks. We then delve into the potential future directions behind the emergence of reasoning abilities within foundation models. We also discuss the relevance of multimodal learning, autonomous agents, and super alignment in the context of reasoning. By discussing these future research directions, we hope to inspire researchers in their exploration of this field, stimulate further advancements in reasoning with foundation models, and contribute to the development of AGI.