Carlo Luschi
6 papers · 2020–2025 · 3 venues · across top CS/AI venues
Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.
Achievements
Jump to papers ↓
🌍
Conference Polyglot · Bronze
🏃
Academic Marathon · Sprint
🌉
Interdisciplinary Bridge · Bronze
Venues
Top co-authors
Research topics
Keywords
low-precision training
(2)
unit scaling
(2)
batch normalization
(1)
neural network optimization
(1)
parallel optimization
(1)
random projection
(1)
quantum chemistry
(1)
foundation model
(1)
density functional theory
(1)
layer normalization
(1)
hyperparameter scaling
(1)
instance normalization
(1)
neural network
(1)
group normalization
(1)
model numerics
(1)
data transfer
(1)
subspace optimization
(1)
llm inference
(1)
attention layer
(1)
intelligence processing unit
(1)
Papers
u-μP: The Unit-Scaled Maximal Update Parametrization
ICLR 2025
SparQ Attention: Bandwidth-Efficient LLM Inference
ICML 2024
Generating QM1B with PySCF_IPU
NeurIPS 2023
Unit Scaling: Out-of-the-Box Low-Precision Training
ICML 2023
Proxy-Normalizing Activations to Match Batch Normalization while Removing Batch Dependence
NeurIPS 2021
Improving Neural Network Training in Low Dimensional Random Bases
NeurIPS 2020