LabFactory builds executable AI labs for diverse scientific tasks
LabFactory: Building and Evaluating Executable AI Labs
Machine Learning
Summary
Building AI systems for scientific problems usually needs lots of steps like gathering data and training models. The authors created LabFactory, a system where an AI builder turns scientific goals into ready-to-run AI labs that can solve specific tasks. These labs combine tools, models, and data resources, and can be tested independently on new data to check their performance. The authors showed that LabFactory can build working labs across different science areas, like predicting molecules and supporting clinical decisions.
What this means in practice
- •For biomedical data teams: Build and deploy task-specific AI systems that integrate models and tools for various biomedical data analysis tasks with automatic evaluation.
- •For software developers in data science: Develop and package AI labs into reusable software artifacts that can be tested and inspected independently on new inputs to ensure reliable performance.
Authors
Jinge Wu, Hongjian Zhou, Mingde Zeng, Jiayuan Zhu, Junde Wu, Jiazhen Pan, Lei Clifton, Andrew Liu, David A. Clifton
Abstract
Scientific tasks specify a desired capability, but realizing it often requires building a computational system tailored to the task---acquiring data, designing representations, training models, implementing tools, and deciding how they are used at inference. We present LabFactory, a framework in which an AI builder turns a scientific brief into an executable AI lab: a task-specific solver that integrates models, knowledge resources, tools, and a controller behind a fixed interface. The builder develops and packages the lab in a metered workspace; a separate host then executes the delivered artifact on held-out inputs, with reference labels kept outside the solver's input interface, and scores its outputs under the task's protocol. This makes the delivered system, rather than the builder's account of its progress, the object of evaluation. We document 28 selected constructions across seven scientific task categories---from molecular and genomic prediction to physiological signals, clinical decision support, and biomedical text---whose delivered labs exceeded their configured reference values on all 33 subtests under host-side execution. Ten contain predictive models fitted during construction; the others assemble retrieval systems, executable analysis environments, and tool-driven workflows around a fixed platform LLM. Together they show that an AI agent can carry a scientific brief all the way to a working lab that can still be invoked, inspected, and checked after construction ends.