NBSR frames neural inference as conjugate Dirichlet evidence accumulation over a DAG with Gumbel-Softmax routing, yielding monotonic precision growth and uncertainty-aware early exit.
Dropout as a bayesian approximation: representing model uncertainty in deep learning
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
-
Neural Bayesian Sequential Routing
NBSR frames neural inference as conjugate Dirichlet evidence accumulation over a DAG with Gumbel-Softmax routing, yielding monotonic precision growth and uncertainty-aware early exit.