Jan Leike
Scholar

Jan Leike

Google Scholar ID: beiWcokAAAAJ
Anthropic
reinforcement learningdeep learningagent alignment
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
57,951
 
H-index
28
 
i10-index
37
 
Publications
20
 
Co-authors
20
list available
Contact
No contact links provided.
Publications
9 items
Constitutional Classifiers++: Efficient Production-Grade Defenses against Universal Jailbreaks
arXiv.org · 2026
Cited
2
Excess Description Length of Learning Generalizable Predictors
arXiv.org · 2026
Cited
0
Natural Emergent Misalignment from Reward Hacking in Production RL
2025
Cited
0
Limit-Computable Grains of Truth for Arbitrary Computable Extensive-Form (Un)Known Games
2025
Cited
0
Unsupervised Elicitation of Language Models
2025
Cited
0
Reasoning Models Don't Always Say What They Think
2025
Cited
7
Auditing language models for hidden objectives
2025
Cited
0
Forecasting Rare Language Model Behaviors
2025
Cited
0
Resume (English only)
Co-authors
20 total
Jeffrey Wu
Jeffrey Wu
Anthropic AI, OpenAI
Co-author 2
Co-author 2
John Schulman
John Schulman
Thinking Machines
Co-author 4
Co-author 4
Co-author 5
Co-author 5
Co-author 6
Co-author 6
Marcus Hutter
Marcus Hutter
Researcher@DeepMind & Professor at ANU
David Scott Krueger
David Scott Krueger
Assistant Professor, University of Montreal, Mila