AIAI researcher Herbie Bradley says leading closed labs use strict internal evals to avoid public benchmark overfittingThe debate centers on labs optimizing models to score high on leaderboards.digg it—3 Sources, 106d ago, first seen 106d agoCombined views115.2K3 Sources, first seen 106d ago636 likesCombined views115.2K3 Sources, first seen 106d ago636 likes
