EFA sets the natural parameters of exponential family distributions to self-attention-transformed context embeddings, subsuming linear latent factor models and beating them on held-out prediction for several datasets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Exponential Family Attention
EFA sets the natural parameters of exponential family distributions to self-attention-transformed context embeddings, subsuming linear latent factor models and beating them on held-out prediction for several datasets.