The Counterfactual Generator — rewrites the field's history without the most-cited papers. Traces how methods, benchmarks, and conclusions would differ if key results had gone the other way.