Bernhard Schölkopf

Research focus

Data Curation & Synthetic Data (1)Open-Source Models & Weights (1)Constitutional AI & AI Ethics (1)Interpretability & Mechanistic Interp (1)

Frequent co-authors

Adrian Hayler (1)A. Hayler (1)Klemens Flöge (1)Alan Arazi (1)

Papers (2)

Aug 17, 2026

Prior Labs1w ago·also Meta AI, NVIDIA, Stanford HAI, Facebook +2

Advancing Open and Reproducible Relational Learning: RelArena-$\alpha$, TabPFN-Rel and RPI

TabPFN-Rel not only tops the leaderboard on RelArena-$\alpha$ but also challenges the notion that specialized architectures are always superior to flattened relational databases.

Adrian Hayler, A. Hayler, Klemens Flöge +56

Data Curation & Synthetic Data Open-Source Models & Weights

Jul 20, 2026

Prakhar Gupta +4Jul 20, 2026

How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?

Alignment tuning installs distinct bias directions in LLMs, allowing for targeted debiasing that recovers unbiased answers while maintaining performance.

Prakhar Gupta, Terry Jingchen Zhang, Florent Draye +2

Constitutional AI & AI Ethics Interpretability & Mechanistic Interp

Search

Bernhard Schölkopf

Publication activitypapers/week, last 8 weeks

Research focus

Frequent co-authors

Papers (2)