Preprints|Review ArticleNot peer reviewedA Taxonomy of Evaluation Leakage in Language-Model BenchmarksNadia Haddad, Ingrid Novak, Kofi MensahVersion 2 (updated) · 14 October 2026 · 10.xxxx/ctm.pp.2026.0004Abstract PDF 3,350 1,4105 cited
Publishing|Research ArticlePeer reviewedRefusal Calibration in Instruction-Tuned Language Models: When should a model decline, and how well do current models know?Nadia Haddad, Hana KovácsVersion 2 · 3 reviewer reports · 9 July 2026 · 10.xxxx/ctm.pb.2026.0007Abstract PDF 3,180 1,2604 cited
