program-evolution

FunL2O presents a unified framework that uses LLM-driven program evolution to automate feature function design for learning-to-optimize methods, showing consistent improvements over hand-crafted features across continuous and discrete optimization tasks.

The paper introduces MILP-Evo, a closed-loop framework that uses LLM-guided program evolution to automatically design white-box MILP solver components (cut selectors and branching rules) by iteratively generating and evaluating candidate programs via end-to-end solver performance on MILP instances.

SMCEvolve introduces a principled framework for LLM-driven program evolution by reformulating it as sampling from a reward-tilted distribution using Sequential Monte Carlo. It provides convergence guarantees and outperforms existing methods across multiple scientific discovery benchmarks.