GitHub - microsoft/stop: Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation :marseysnappy: :marseysuspicious: Surely Microsoft is competent enough to make this work?

Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation

This is the repo for the paper: Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation

@article{zelikman2023self, title={Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation}, author={Eric Zelikman, Eliana Lorch, Lester Mackey, Adam Tauman Kalai}, journal={arXiv preprint arXiv:2310.02304}, year={2023} }

Abstract: Several recent advances in AI systems (e.g., Tree-of-Thoughts and Program-Aided Language Models) solve problems by providing a "scaffolding" program that structures multiple calls to language models to generate better outputs. A scaffolding program is written in a programming language such as Python. In this work, we use a language-model-infused scaffolding program to improve itself. We start with a seed "improver" that improves an input program according to a given utility function by querying a language model several times and returning the best solution. We then run this seed improver to improve itself. Across a small set of downstream tasks, the resulting improved improver generates programs with significantly better performance than its seed improver. Afterward, we analyze the variety of self-improvement strategies proposed by the language model, including beam search, genetic algorithms, and simulated annealing. Since the language models themselves are not altered, this is not full recursive self-improvement. Nonetheless, it demonstrates that a modern language model, GPT-4 in our proof-of-concept experiments, is capable of writing code that can call itself to improve itself. We critically consider concerns around the development of self-improving technologies and evaluate the frequency with which the generated code bypasses a sandbox.