For uniform keys on the d-dimensional sphere, softmax attention becomes selective at inverse temperature scaling β_n* ≍ n^{2/(d-1)}, with explicit limiting laws for attention weights and outputs in each regime.
Physics reports , volume=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Derives exact closed-form effect of adversarial kicks on the order parameter in Kuramoto networks, showing coupling-independent increments even at weak synchronization and explaining amplification near criticality.
citing papers explorer
-
Scaling Limits of Long-Context Transformers
For uniform keys on the d-dimensional sphere, softmax attention becomes selective at inverse temperature scaling β_n* ≍ n^{2/(d-1)}, with explicit limiting laws for attention weights and outputs in each regime.
-
Analytical foundation for adversarial synchronization control in oscillator networks
Derives exact closed-form effect of adversarial kicks on the order parameter in Kuramoto networks, showing coupling-independent increments even at weak synchronization and explaining amplification near criticality.