Conversation
…toolkit optimization/athena.py implements the closed loop of GRAFT-ATHENA (arXiv 2605.11117) over evolve_lineage: a strategist agent writes methods as annotated callables over a typed 1-D finite-difference and spectral toolkit, an evaluator measures failure modes and convergence class against exact references kept out of the prompt, and a type-directed memory seeds each problem. Tested on the paper's spectral Burgers case (Section 2.8), which both warm and cold arms reconstruct.
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR adds an example derived from GRAFT-ATHENA, a state-of-the-art LLM-based optimizer for scientific computing problems, in
docs/source/llm_examples/optimization/athena.py. The original is incredibly convoluted and almost impossible to understand or reproduce, for reasons that seem entirely unrelated to its effectiveness.This replaces almost all of the bespoke machinery from the paper with ordinary typed code generation and simplifies it to its essence as an incremental improvement over the AVO example
docs/source/llm_examples/optimization/avo.pyto use learning (in-context learning, in this case) to warm-start search on new problems. The example script reproduces a version of the original paper's spectral Burgers case study (Section 2.8); reproducing the other case studies would be straightforward.