2 citations · 2 across the 2 of their papers we have counts for
3 papers
stat.ML2026
Identifiability Without Gaussianity: Symbolic World Models and Near-Infinite Temporal Consistency
Seth Dobrin, Åukasz Chmiel
Klindt, LeCun, and Balestriero (arXiv:2605.26379) proved that Joint-Embedding Predictive Architectures (JEPAs) achieve linear identifiability, the linear recovery of the world's tr…
cs.AI2026★ 2 cited
The Unfireable Safety Kernel: Execution-Time AI Alignment for AI Agents and Other Escapable AI Systems
Seth Dobrin, Åukasz Chmiel
AI agents are granted access to tools, APIs, and other infrastructure, making them active principals in those systems. The dominant approach places controls inside the agent's own…
cs.AI2026
ARYA: A Physics-Constrained Composable & Deterministic World Model Architecture
Seth Dobrin, Lukasz Chmiel
This paper presents ARYA, a composable, physics-constrained, deterministic world model architecture built on five foundational principles: nano models, composability, causal reason…