Shai, Adam S.

1. Transformers represent belief state geometry in their residual stream
   2024/05/24 by Adam S. Shai, Sarah E. Marzen, Sarah Marzen +9 · 1 voice · 17 citations
   Computer Science · Engineering · Mathematics · #Algorithm #Computation and Language (cs.CL) #Computer science #Data Stream Mining Techniques #Electrical engineering #Engineering #FOS: Computer and information sciences #Geometry #Machine Learning (cs.LG) #Machine Learning and Algorithms #Mathematics #Residual #State (computer science) #Transformer #Voltage #cs.CL #cs.LG
2. Constrained belief updates explain geometric structures in transformer representations
   2025/02/04 by Mateusz Piotrowski, Paul M. Riechers, Piotrowski, Mateusz +5 · 1 voice · 5 citations
   Computer Science · #Rough Sets and Fuzzy Logic #cs.LG