analiza aiqbench
ApexQuant introduces a calibration-free quantization technique that recursively refines residual errors using random rotations to achieve near full-precision accuracy without requiring training data. This approach is particularly relevant for deploying large language models in privacy-sensitive domains like healthcare and earth observation where ground-truth data is inaccessible. By enabling progressive compression into valid lower-bit models, the method advances the practical efficiency of AI inference under strict data constraints.
ApexQuant: Data-Free Elastic Quantization by Residual Re-Isotropization
arXiv:2610.07904v1 Announce Type: new Abstract: We introduce ApexQuant, a calibration-free quantization method that recursively re-quantizes the residual error, serving as a refinement layer on top of existing quantizers. We establish that a fresh random rotation returns each residual to the uniform distribution on the hypersphere, which characterizes the rate of progressive error decay across successive passes. This result lets us determine, before any weight is read, how many passes a layer needs for a target weight-space error. Every prefix is itself a valid lower-rate model, so one artifact serves several precisions. We instantiate ApexQuant with three interchangeable stages, scalar, $E_8$ and trellis, and validate it on four open-weight LLMs and on Earth-observation and medical domains where in-distribution data is often unattainable as imagery arrives under restrictive licences or due to patient material under privacy constraints. Progressive re-isotropization comes within a few percent of full precision at four bits and gives the best two-bit arm we measure, in a completely data-free setting.
Abstract:We introduce ApexQuant, a calibration-free quantization method that recursively re-quantizes the residual error, serving as a refinement layer on top of existing quantizers. We establish that a fresh random rotation returns each residual to the uniform distribution on the hypersphere, which characterizes the rate of progressive error decay across successive passes. This result lets us determine, before any weight is read, how many passes a layer needs for a target weight-space error. Every prefix is itself a valid lower-rate model, so one artifact serves several precisions. We instantiate ApexQuant with three interchangeable stages, scalar, $E_8$ and trellis, and validate it on four open-weight LLMs and on Earth-observation and medical domains where in-distribution data is often unattainable as imagery arrives under restrictive licences or due to patient material under privacy constraints. Progressive re-isotropization comes within a few percent of full precision at four bits and gives the best two-bit arm we measure, in a completely data-free setting.
Machine Learning (cs.LG)
arXiv:2610.07904 [cs.LG]
(or arXiv:2610.07904v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.07904 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 250 cuvinte
Subiecte conexe
Hyperelastic constitutive model discovery with differentiable finite elements and structure-preserving neural networks
arXiv · 9 sept. 2026
MRVQ: One Resident Index for Dimension- and Rate-Elastic Vector Search
arXiv · 6 oct. 2026
Correcting Neural Operator Spectral Bias via Diffusion Posterior Sampling with Sparse Observations
arXiv · 3 iun. 2026
[AINews] Meta Muse Wonder & Wonder Full: MSL’s Claude 5.5-tier Frontier Model
Latent Space · 13 oct. 2026