SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.02304 · 2023

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

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 7 functions out of this paper's own repositories and ran 4 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
microsoft/stop canonical 4 of 7
FunctionStatusWhere it lives
algorithm Ran microsoft/stop/tasks/maxcut/seed_algorithm.py
code served (permissive licence) · get_code("948a2eb852c7d75e")
algorithm Ran microsoft/stop/tasks/modified_quadratic_assignment/seed_algorithm.py
code served (permissive licence) · get_code("7e0ada009b7cf021")
algorithm Ran microsoft/stop/tasks/three_sat/seed_algorithm.py
code served (permissive licence) · get_code("4ba2a1030ad49f51")
random_walk_solver Ran microsoft/stop/tasks/three_sat/seed_algorithm.py
code served (permissive licence) · get_code("bdcd923bdde3921f")
extract_code Not yet run microsoft/stop/helpers.py
code served (permissive licence) · get_code("dc62532cec8323f5")
find_largest_code_block_line_by_line Not yet run microsoft/stop/helpers.py
code served (permissive licence) · get_code("d7a13e94854d7e2c")
run_by_creating_file Not yet run microsoft/stop/helpers.py
code served (permissive licence) · get_code("d9043050b18efcd6")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Several recent advances in AI systems solve problems by providing a "scaffolding" program that structures multiple calls to language models (LMs) 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 an LM 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. A variety of self-improvement strategies are 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 experiments, is capable of writing code that can call itself to improve itself. We consider concerns around the development of self-improving technologies and evaluate the frequency with which the generated code bypasses a sandbox.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2310.02304")
get_code_for_paper("2310.02304")
have("2310.02304")

Connect an agent — have() is free.