jingyongai/

executable-knowledge-lifecycles-benchmark

This dataset is the frozen semantic benchmark for Experiments 1A and 1B of Executable Knowledge Lifecycles as Dynamical Control in Agent Systems. It tests whether an untrusted natural-language parser can be placed behind a typed, model-external compilation boundary without acquiring validation authority.

Version: 0.2.6-exp.1-final Version DOI: 10.5281/zenodo.21361204 Source release: GitHub v0.2.6-exp.1-final Release collection:… See the full description on the dataset page: https://huggingface.co/datasets/jingyongai/executable-knowledge-lifecycles-benchmark.