From Language Models to Practical Self-Improving Computer Agents
This work addresses the challenge of automating tool development for AI agents, potentially reducing human effort in enhancing LLM capabilities, though it appears incremental as it builds on existing non-parametric augmentation methods.
The authors tackled the problem of creating AI computer agents that can autonomously self-improve by generating software tools, enabling them to solve increasingly complex tasks without manual human engineering. They demonstrated through case studies that a minimal querying loop with prompt engineering allows an LLM agent to augment itself with capabilities like retrieval, internet search, and web navigation, effectively solving real-world computer tasks such as automated software development.
We develop a simple and straightforward methodology to create AI computer agents that can carry out diverse computer tasks and self-improve by developing tools and augmentations to enable themselves to solve increasingly complex tasks. As large language models (LLMs) have been shown to benefit from non-parametric augmentations, a significant body of recent work has focused on developing software that augments LLMs with various capabilities. Rather than manually developing static software to augment LLMs through human engineering effort, we propose that an LLM agent can systematically generate software to augment itself. We show, through a few case studies, that a minimal querying loop with appropriate prompt engineering allows an LLM to generate and use various augmentations, freely extending its own capabilities to carry out real-world computer tasks. Starting with only terminal access, we prompt an LLM agent to augment itself with retrieval, internet search, web navigation, and text editor capabilities. The agent effectively uses these various tools to solve problems including automated software development and web-based tasks.