Scholar
Tomek Korbak
Google Scholar ID: YQ5rrk4AAAAJ
UK AI Security Institute
Citations & Impact
Citations
4,090
H-index
22
i10-index
29
Publications
20
Co-authors
8
list available
Publications
Resume (English only)
Academic Achievements
- Published multiple papers at top-tier conferences including ICLR, ICML, NeurIPS, and COLM, such as:
- “A sketch of an AI control safety case”
- “Looking Inward: Language Models Can Learn About Themselves by Introspection” (ICLR 2025)
- “Is model collapse inevitable? breaking the curse of recursion by accumulating real and synthetic data” (COLM 2024)
- “Pretraining Language Models with Human Preferences” (ICML 2023, oral)
- “On reinforcement learning and distribution matching for fine-tuning language models with no catastrophic forgetting” (NeurIPS 2022, oral)
- Many papers accompanied by open-source code.
