Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Overlapping Schwarz Attention: Hierarchical Attention via Domain Decomposition
Stephan Köhler, Oliver Rheinbach
We propose a hierarchical attention mechanism based on two-level overlapping Schwarz domain decomposition. The method is motivated by domain decomposition methods in partial differ…
cs.LG2026
How Token Influence Decays with Distance: A Green-Function View of Trained Language Models
Matthias Brändel, Stephan Köhler, Oliver Rheinbach
We study how the next-token prediction of an autoregressive Transformer language model changes under small perturbations of earlier input token embeddings. Motivated by operator le…