Tag
This paper systematically evaluates reject inference methods in credit scoring and identifies a failure mode where accuracy improves while recall collapses, creating an illusion of improvement while rejection quality deteriorates. It proposes a controlled exploration strategy that breaks the feedback loop and shows that even minimal exploration rates are sufficient to diagnose the problem.
This paper develops a framework to grade the capability to infer in data-driven systems under the European AI Act, using credit scoring as a case study to illustrate where inference occurs and where regulatory clarity is needed.
This survey examines computational nondeterminism in financial AI systems, covering tabular models, graph networks, and LLM-based workflows, and proposes a layered evaluation framework for auditability.