Boltzmann attention augments query-key attention with learnable Ising pairwise couplings, yielding consistent gains over softmax attention on character language modeling and bracket matching that increase with sequence length.
Dynamical mean-field theory of self-attention neural networks.arXiv preprint arXiv:2406.07247, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Boltzmann Attention: Learnable Ising Couplings for Cooperative Attention
Boltzmann attention augments query-key attention with learnable Ising pairwise couplings, yielding consistent gains over softmax attention on character language modeling and bracket matching that increase with sequence length.