← Back to The Lab
Dr. Aris Thorne
Lead Scientist, NLP
Cohere
LLM Scaling
Model Behavior
Alignment
About
Dr. Aris Thorne is a Lead Scientist focused on large-scale language modeling and alignment. His research spans scaling laws, emergent behaviors, and the mechanistic interpretability of transformer-based systems. He has mentored over 50 researchers across academia and industry, and frequently collaborates with open-science groups on reproducibility studies.
Education & Background
- •Ph.D. in Computer Science, Stanford University
- •M.S. in Machine Learning, Carnegie Mellon University
- •B.S. in Mathematics, MIT
Active RFPs
Published Papers
- Scaling laws for instruction-tuned models— NeurIPS 2024
- On the alignment tax of RLHF— ICML 2023
- Probing emergent reasoning in 70B models— ACL 2023