Our paper "An Empirical Investigation of Bias and Unfairness in Case-Based Adaptation" was awarded with the Best Paper Award of this year's International Conference on Case-Based Reasoning (ICCBR 2026). This conference is the premier annual meeting of the case-based reasoning (CBR) community and the leading international conference on this topic. Case-based reasoning (CBR) is an AI problem-solving approach that reuses solutions from past similar cases to address new problems. With an acceptance rate of 18.3% (main track) and 10.2% (oral), the ICCBR is a highly selective conference.

Our award-winning paper explores how bias and unfairnessfrom the real world can seep into AI systems, specifically in Case-Based Reasoning, where solutions are derived from similar past cases. While prior work has examined fairness in data retrieval and storage, we focus on the reuse phase: How do adaptation methods in k-nearest neighbor classification influence fairness in decisions?
We demonstrate that even seemingly neutral algorithms - such as those used in credit approval or risk assessment - can amplify social inequalities by systematically disadvantaging certain groups. Our research fills a critical gap in understanding fairness in AI and paves the way for more equitable and transparent systems.

On the occasion of receiving the Best Paper Award, we are releasing an Editio Emendata version of the official paper here, which corrects a few typographical errors that were discovered after the final print approval. There are no differences in content.