A hyper-network that predicts segmentation UNet weights from per-sample learned latent vectors yields a structured map of annotation styles and a way to flag erroneous labels.
Adaptive Computation and Ma- chine Learning series, MIT Press (2024)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
HyperSORT: Self-Organising Robust Training with hyper-networks
A hyper-network that predicts segmentation UNet weights from per-sample learned latent vectors yields a structured map of annotation styles and a way to flag erroneous labels.