Gennady Pekhimenko
7 papers · 2021–2025 · 5 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+2 more ↓ Show less ↑
π Interdisciplinary Bridge π§ Keyword Pioneer π Conference Polyglot (5) π Cross-Pollinator (5) π Renaissance Researcher (5)
πΊοΈ
Taxonomy Completionist
(16)
π£
Hot Topic Early Bird
Conferences
NIPS (3)
ACL (1)
ICLR (1)
ICML (1)
OSDI (1)
Top co-authors
Keywords
stochastic gradient descent
(1)
efficient training
(1)
neural network training
(1)
prompt engineering
(1)
distributed learning
(1)
model pretraining
(1)
distributed optimization
(1)
model parallelism
(1)
communication efficiency
(1)
memory optimization
(1)
transformer training
(1)
performance model
(1)
tensor compilation
(1)
kernel generation
(1)
tile abstraction
(1)
search space
(1)
decentralized training
(1)
gradient accumulation
(1)
programming language
(1)
collaborative training
(1)
Papers
APPL: A Prompt Programming Language for Harmonious Integration of Programs and Large Language Model Prompts
ACL 2025
TypyBench: Evaluating LLM Type Inference for Untyped Python Repositories
ICML 2025
Guaranteed Approximation Bounds for Mixed-Precision Neural Operators
ICLR 2024
Tempo: Accelerating Transformer-Based Model Training through Memory Footprint Reduction
NIPS 2022
ROLLER: Fast and Efficient Tensor Compilation for Deep Learning
OSDI 2022
Distributed Deep Learning In Open Collaborations
NIPS 2021
Moshpit SGD: Communication-Efficient Decentralized Training on Heterogeneous Unreliable Devices
NIPS 2021