What makes LLM Agents

Lilian Weng's blog on ‘LLM Powered Autonomous Agents’ is one of the most concise and well-structured overview of Agentic behaviour and automation achieved through Large language models. While building agents using open source frameworks and experimenting to improve their reasoning and generation, I came across this blog. It inspired me to dig deeper into each component and how to improve them. I want to start a series where I explore the foundational as well as SoTA research in this field and write my finding.

I will start with sharing some resources that are useful and some I will be using for this series:
1. LLM Powered Autonomous Agents - https://lilianweng.github.io/posts/2023-06-23-agent/
2. System 2 research - https://github.com/open-thought/system-2-research
3. LLM-Agents-Papers - https://github.com/AGI-Edgerunners/LLM-Agents-Papers