Scholar
Rob Brekelmans
Google Scholar ID: M6ADg_UAAAAJ
Vector Institute
Citations & Impact
Citations
770
H-index
15
i10-index
16
Publications
20
Co-authors
0
Publications
Resume
Academic Achievements
- Winner of Best Paper Award at ICML 2024 for the work 'Probabilistic Inference in Language Models via Twisted Sequential Monte Carlo'; Published research on Variational Representations of Annealing Paths in the Information Geometry journal; Published Extended State Space Importance Sampling (applied to mutual information estimation) at ICLR 2022.
Research Experience
- Postdoctoral Fellow at the Vector Institute in Toronto, working with Alireza Makhzani and Roger Grosse; Interned with the AI Safety Analysis team at DeepMind, working with Pedro Ortega and Tim Genewein.
Education
- PhD from University of Southern California in 2022, working with Greg Ver Steeg and Aram Galstyan.
Background
- Research interests include control-as-inference perspectives across recent language and diffusion modeling settings. Currently a Postdoctoral Fellow at the Vector Institute in Toronto.
Miscellany
- Wrote a blog post reviewing control-as-inference perspectives.