Artur B. Carneiro
AI researcher at Stanford. Working on efficient reasoning, mechanistic interpretability, and making large models smaller and faster.
Writing
Scaling Sparse Attention for Long-Context ReasoningMar 2026On the Geometry of Representation Collapse in LLMsJan 2026Why Reward Models Plateau — and What To Do About ItNov 2025A Primer on Mechanistic InterpretabilityAug 2025Notes on Diffusion Models as World SimulatorsMay 2025View all →