3 papers
cs.LG2026
Incoherence in Goal-Conditioned Autoregressive Models
Jacek Karwowski, Raymond Douglas
We investigate mathematically the notion of incoherence: a structural issue with reinforcement learning policies derived by naive goal-conditioning of autoregressive models. We foc…
cs.CG2025
Hilbert geometry of the symmetric positive-definite bicone: Application to the geometry of the extended Gaussian family
Jacek Karwowski, Frank Nielsen
The extended Gaussian family is the closure of the Gaussian family obtained by completing the Gaussian family with the counterpart elements induced by degenerate covariance or dege…
cs.LG2024
Characterizing stable regions in the residual stream of LLMs
Jett Janiak, Jacek Karwowski, Chatrik Singh Mangat +3
We identify stable regions in the residual stream of Transformers, where the model's output remains insensitive to small activation changes, but exhibits high sensitivity at region…