Gödel Agent: A self-referential agent framework for recursive self-improvement

heh I was just working on something that tries to improve itself today. I wrote a simple agent executor that makes calling one a simple function call, and then wrote an agent which invents other agents. By calling that in a loop for a while I ended up with, effectively, a large library of functions I not only didn&#x27;t write but didn&#x27;t even think up.<p>By passing those functions as tools in LLM requests any of the agents can make use of any of the other agents so it&#x27;s basically expanding its own capabilities.<p>Not quite sure what task to sick it on yet but it&#x27;s fun to play with.

Shameless plug, for anyone who&#x27;s interested in &quot;self-improvement&quot; agent check out StreamBench[1] where we benchmark and try out what&#x27;s essential for improvements in online settings. Basically we find feedback signal is vital and the stronger the signal the more improvement you can get if you were able to feed it back to the agent in terms of weights (LoRA) or in-context examples.<p>[1] <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;2406.08747" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;2406.08747</a>

Let’s see the code. A bit skeptical, this hasnt over complicated something architecturally. Need more clear drawings of architecture. What prompts exist, what tool calls are made, and what gets updated.

Can it modify its training data?

&gt; For the Godel Agent, we utilize the “gpt-4o-2024-05-13” model (OpenAI et al., 2024), whereas the optimized policy and baseline models are evaluated using the “gpt-3.5-turbo-0125” model (OpenAI, 2022) to reduce computational costs and ensure a fair comparison.<p>Doesn&#x27;t seem fair at all.

I’m skeptical this would work in production better than RLHF, if the agent makes a mistake, how is it supposed to know to correct itself and understand what it did wrong to prevent it? It seems better to try again recursively until it finds the solution like a human

That’s a lot of words, where is the code to reproduce?

What a strange loop

&gt;The rapid advancement of large language models (LLMs) has significantly enhanced the capabilities of AI-driven agents across various tasks<p>No it hasn&#x27;t.

If their demo work, they must be close to AGI right?

meh im not convinced that any sort of framework or side tool that works on top of large language models is the solution<p>we really need something intelligent (no, o1 doesn&#x27;t count) and its unclear what that will look like. Perhaps it will be some RNN with neurosymbolism