LLM-based Agentic Reasoning Frameworks: A Survey from Methods to Scenarios
This work addresses the need for a structured understanding of LLM-based agentic reasoning frameworks for researchers and practitioners in AI, but it is incremental as it synthesizes existing methods rather than introducing new ones.
This survey tackles the problem of understanding and categorizing the diverse reasoning frameworks used in LLM-based agent systems by proposing a systematic taxonomy and analyzing their applications across various scenarios, such as scientific discovery and healthcare, to provide a comprehensive overview for the research community.
Recent advances in the intrinsic reasoning capabilities of large language models (LLMs) have given rise to LLM-based agent systems that exhibit near-human performance on a variety of automated tasks. However, although these systems share similarities in terms of their use of LLMs, different reasoning frameworks of the agent system steer and organize the reasoning process in different ways. In this survey, we propose a systematic taxonomy that decomposes agentic reasoning frameworks and analyze how these frameworks dominate framework-level reasoning by comparing their applications across different scenarios. Specifically, we propose an unified formal language to further classify agentic reasoning systems into single-agent methods, tool-based methods, and multi-agent methods. After that, we provide a comprehensive review of their key application scenarios in scientific discovery, healthcare, software engineering, social simulation, and economics. We also analyze the characteristic features of each framework and summarize different evaluation strategies. Our survey aims to provide the research community with a panoramic view to facilitate understanding of the strengths, suitable scenarios, and evaluation practices of different agentic reasoning frameworks.