Pion is an optimizer that preserves the singular values of weight matrices in LLM training by applying orthogonal equivalence transformations.
Efficient riemannian optimization on the stiefel manifold via the cayley transform
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
representative citing papers
ReSpinQuant achieves state-of-the-art accuracy in W4A4 and W3A3 LLM quantization by using efficient residual subspace rotation approximations that match layer-wise performance while retaining the inference speed of global rotation methods.
citing papers explorer
-
Pion: A Spectrum-Preserving Optimizer via Orthogonal Equivalence Transformation
Pion is an optimizer that preserves the singular values of weight matrices in LLM training by applying orthogonal equivalence transformations.
-
ReSpinQuant: Efficient Layer-Wise LLM Quantization via Subspace Residual Rotation Approximation
ReSpinQuant achieves state-of-the-art accuracy in W4A4 and W3A3 LLM quantization by using efficient residual subspace rotation approximations that match layer-wise performance while retaining the inference speed of global rotation methods.
- Tensor-Programmable Quantum Circuits for Solving Differential Equations