Joeran Beel

Please visit https://isg.beel.org/people/joeran-beel/ for more details about me.

A RecSys Paper for $20: Why We Must Build, Evaluate, and Govern Autonomous Recommender Systems Research Labs (AutoRecLabs)

Joeran Beel1,2, Bela Gipp3, Tobias Vente4, Moritz Baumgart1,3, Philipp Meister3 and Sinan Pourazari1 1 University of Siegen, Siegen, Germany2 Recommender-Systems.com, Düsseldorf, Germany3 University of Göttingen, Göttingen, Germany4 University of Antwerp, Antwerp, Belgium FRAME’26: Methodology First – Rethinking Research Assessment in RecSys Workshop, September 28, 2026, Minneapolis, Minnesota, USA. Abstract Recommender Read more…

By Joeran Beel, ago

“If AI can run an AI or RecSys experiment for $1, why spend four years doing a PhD?” — Keynote and Mentor at the ACM RecSys Doctoral Consortium 2026

My keynote at the Doctoral Consortium of ACM RecSys 2026 in Minneapolis On 27 September 2026, I gave a keynote at the Doctoral Consortium of ACM RecSys 2026 in Minneapolis. The title was deliberately provocative: “When AI Does RecSys Experiments for $1: Why Spend Four Years on a PhD?” The Read more…

By Joeran Beel, ago

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…

By Joeran Beel, ago

AutoRecLab: An Autonomous Recommender Systems Lab

Abstract. Recommender systems (RecSys) research depends on extensive empirical evaluation, yet translating experimental designs into executable code remains a manual, error-prone process. This paper presents AutoRecLab, a Python-based autonomous RecSys lab that automates RecSys experimentation from natural-language prompts. Starting from a research idea, AutoRecLab derives explicit experiment requirements, develops and Read more…

By Joeran Beel, ago

From AutoRecSys to AutoRecLab: A Call to Build, Evaluate, and Govern Autonomous Recommender-Systems Research Labs

Joeran Beel (University of Siegen, Germany) Bela Gipp (University of Göttingen, Germany) Tobias Vente (University of Antwerp, Belgium) Moritz Baumgart (University of Siegen, Germany) Philipp Meister (University of Göttingen, Germany) Pre-Print @misc{beel2025autorecsysautoreclabbuildevaluate, title={From AutoRecSys to AutoRecLab: A Call to Build, Evaluate, and Govern Autonomous Recommender-Systems Research Labs}, author={Joeran Beel and Read more…

By Joeran Beel, ago

Min-Yen Kan and Joeran Beel speak with ‘Nature’ about the potential (and threats) of AI Scientists

Following my recent visit to Prof. Min-Yen Kan at the National University of Singapore, Moritz Baumgart, Min-Yen Kan and I co-authored a paper critically evaluating Sakana’s AI Scientist (to be published in SIGIR Forum). Our review attracted the attention of Nature, which was preparing an editorial feature on the opportunities Read more…

By Joeran Beel, ago

Reproducing Machine Learning: Seven Bachelor Projects That Hold Science Accountable

Over the past semester, several Bachelor students at the University of Siegen undertook a bold challenge in our Machine Learning Praktikum: they didn’t just learn algorithms—they tried to reproduce published machine learning or recommender systems research. Each student or team selected a recent paper, rebuilt the experimental pipeline, validated (or Read more…

By Joeran Beel, ago

Versorgungsauskunft für neue Beamte/Professoren: Mein Fall, mein Widerspruch, mein Ergebnis (8 Jahre mehr = 100.000€ mehr Pension)

In diesem Blogpost dokumentiere ich, wie das Landesamt für Besoldung und Versorgung NRW (LBV) versucht hat, mir eine Versorgungsauskunft faktisch zu verwehren, wie anschließend zu meinen Ungunsten entschieden wurde – und wie ich am Ende dennoch rund acht Jahre mehr ruhegehaltsfähige Zeit anerkannt bekommen habe, als zunächst vorgesehen. Das entspricht Read more…

By Joeran Beel, ago

Evaluating Sakana’s AI Scientist: Bold Claims, Mixed Results, and a Promising Future?

Abstract. Recently, Sakana.ai introduced the AI Scientist, which claims to automate the entire research lifecycle and conduct research autonomously. This is a concept we call Artificial Research Intelligence (ARI). Achieving ARI would be a major milestone toward Artificial General Intelligence (AGI) and a prerequisite to achieving Super Intelligence. The AI Read more…

By Joeran Beel, ago

Checky, the Paper-Submission Checklist Generator for Authors, Reviewers and LLMs

Following our proposal for evidence-based best-practices for recommender systems evaluation and our Dagstuhl manuscript about Best-Practices for Offline Evaluations of Recommender Systems, we are glad to announce Checky, a tool for conference chairs and journal editors to create and manage submission checklists. Abstract. Submission checklists have become increasingly prevalent for Read more…

By Joeran Beel, ago

ChatGPT (und ich) verlieren gegen die LVM vor dem Amtsgericht – ein Praxistest für den KI-Anwalt

Vor etwa einem halben Jahr endete mein erster Versuch, mit Unterstützung von ChatGPT einen Zivilprozess selbst zu führen. Das Ergebnis vorweg: Wir haben verloren, auf eine ziemlich unbefriedigende Art. Das Amtsgericht hat nicht entschieden, ob meine tatsächliche Position gegenüber der Beklagten richtig oder falsch war. Es hat meine Klage als Read more…

By Joeran Beel, ago