Your nDCG Score May Be Correct. But the Conclusion? A Call for Stability-Aware Offline Recommender Evaluation

Joeran BeelUniversity of Siegen, Siegen, GermanyRecommender-Systems.com, Düsseldorf, GermanyORCID: 0000-0002-4537-5573 FRAME’26: Methodology First – Rethinking Research Assessment in RecSys Workshop, September 28, 2026, Minneapolis, Minnesota, USA Abstract Offline recommender-system evaluations often end with one selected score per algorithm, such as nDCG. That score may be correct for the reported condition, yet Read more…