Search papers, labs, and topics across Lattice.
This paper introduces LACE, an innovative AutoML framework that leverages large language models to evolve complete executable pipeline programs, moving beyond traditional methods that are limited to predefined components. By maintaining a population of scikit-learn-compatible Python classes and employing an evolutionary loop, LACE enables the generation of customizable and editable machine learning pipelines. Evaluated across 68 OpenML classification tasks, LACE outperforms established systems like auto-sklearn and H2O, while providing a unique advantage of producing reusable code rather than just model objects.
LACE transforms AutoML by enabling the evolution of fully executable Python pipelines, allowing practitioners to directly edit and reuse generated models.
Automated machine learning (AutoML) systems search for pipelines within a space of preprocessing operators, learners, and hyper-parameters specified in advance: they can select and tune known components, but cannot produce structure outside that space. We present LACE, an AutoML framework that instead searches over complete executable pipeline programs: an evolutionary loop maintains a population of scikit-learn-compatible Python classes, and a large language model acts as the variation operator. To our knowledge, LACE is the first to formulate general tabular pipeline AutoML this way, evaluated on standardized OpenML tasks under a leakage-controlled protocol that withholds dataset identity from the generator. Because every candidate is ordinary Python, the returned pipeline and the search that produced it can be inspected and edited directly, rather than only through a framework's model objects. On 68 OpenML classification tasks, LACE with GPT-5.4-mini significantly outperforms auto-sklearn, H2O, and a fixed XGBoost baseline, with no detectable difference against AutoGluon, the strongest search-based system evaluated, while covering the full benchmark. Newer tabular foundation models are more accurate on the subset of tasks they support, but apply a fixed pretrained predictor rather than returning an editable task-specific program. LACE's contribution is therefore not raw accuracy but a search space defined by code: complete coverage, pipelines practitioners can reuse directly, and a component set extended by editing the prompt rather than the framework.
