Conglong Li
7 papers · 2021–2024 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+3 more ↓ Show less ↑
π Interdisciplinary Bridge π§ Keyword Pioneer π Cross-Pollinator (12) πΊοΈ Taxonomy Completionist (19) π Conference Polyglot (4)
π£
Hot Topic Early Bird
π
Grand Slam
π
Keyword Champion
(2)
Conferences
NIPS (3)
ICML (2)
AAAI (1)
ICLR (1)
Top co-authors
Keywords
model compression
(4)
large language model
(3)
knowledge distillation
(2)
training efficiency
(2)
curriculum learning
(1)
model pretraining
(1)
deep learning
(1)
neural network optimization
(1)
efficient computing
(1)
distributed training
(1)
model inference
(1)
sparse model
(1)
adaptive optimization
(1)
adam optimizer
(1)
communication compression
(1)
mixture of expert
(1)
training stability
(1)
data efficiency
(1)
inference optimization
(1)
weight quantization
(1)
Papers
DeepSpeed Data Efficiency: Improving Deep Learning Model Quality and Training Efficiency via Efficient Data Sampling and Routing
AAAI 2024
Maximizing Communication Efficiency for Large-scale Training via 0/1 Adam
ICLR 2023
The Stability-Efficiency Dilemma: Investigating Sequence Length Warmup for Training GPT Models
NIPS 2022
ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers
NIPS 2022
XTC: Extreme Compression for Pre-trained Transformers Made Simple and Efficient
NIPS 2022
DeepSpeed-MoE: Advancing Mixture-of-Experts Inference and Training to Power Next-Generation AI Scale
ICML 2022
1-bit Adam: Communication Efficient Large-Scale Training with Adamβs Convergence Speed
ICML 2021