1 citations · 1 across the 5 of their papers we have counts for
Showing 2025Show all
2 papers · 1 filter
cs.LG2025
The Curved Spacetime of Transformer Architectures
Riccardo Di Sipio, Jairo Diaz-Rodriguez, Luis Serrano
We present a geometric framework for understanding Transformer-based language models, drawing an explicit analogy to General Relativity. Queries and keys induce an effective metric…
cs.CL2025
Rethinking LLM Training through Information Geometry and Quantum Metrics
Riccardo Di Sipio
Optimization in large language models (LLMs) unfolds over high-dimensional parameter spaces with non-Euclidean structure. Information geometry frames this landscape using the Fishe…