B³-Net improves multi-task dense prediction by estimating patch-wise evidence precision, fusing it into a reliability-weighted posterior bridge, and redistributing via bounded updates to limit contamination from unreliable task evidence.
UberNet: Training a universal convolutional neural net- work for low-, mid-, and high-level vision using diverse datasets and limited memory
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
$\mathcal{B}^{3}$-Net: Controlled Posterior Bridge Learning for Multi-Task Dense Prediction
B³-Net improves multi-task dense prediction by estimating patch-wise evidence precision, fusing it into a reliability-weighted posterior bridge, and redistributing via bounded updates to limit contamination from unreliable task evidence.