λ-Orthogonality regularization enables distribution-specific adaptation of representations via affine transformations while retaining original learned structures.
Imagenet large scale visual recognition challenge
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2representative citing papers
MARR uses per-module adaptive residual scaling updated by PID feedback to balance error correction against Hessian-approximation bias in low-bit PTQ.
citing papers explorer
-
$\boldsymbol{\lambda}$-Orthogonality Regularization for Compatible Representation Learning
λ-Orthogonality regularization enables distribution-specific adaptation of representations via affine transformations while retaining original learned structures.
-
MARR: Module-Adaptive Residual Reconstruction for Low-Bit Post-Training Quantization
MARR uses per-module adaptive residual scaling updated by PID feedback to balance error correction against Hessian-approximation bias in low-bit PTQ.