Picture for Zhen Xie

Zhen Xie

CityLLaVA: Efficient Fine-Tuning for VLMs in City Scenario

Add code
May 06, 2024
Viaarxiv icon

DeepSpeed4Science Initiative: Enabling Large-Scale Scientific Discovery through Sophisticated AI System Technologies

Add code
Oct 11, 2023
Viaarxiv icon

A Comprehensive Performance Study of Large Language Models on Novel AI Accelerators

Add code
Oct 06, 2023
Viaarxiv icon

Benchmarking and In-depth Performance Study of Large Language Models on Habana Gaudi Processors

Add code
Sep 29, 2023
Viaarxiv icon

Transfer Learning Across Heterogeneous Features For Efficient Tensor Program Generation

Add code
Apr 11, 2023
Viaarxiv icon

Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation

Add code
Aug 26, 2020
Figure 1 for Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation
Figure 2 for Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation
Figure 3 for Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation
Figure 4 for Adaptive Neural Network-Based Approximation to Accelerate Eulerian Fluid Simulation
Viaarxiv icon

Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation

Add code
Aug 26, 2020
Figure 1 for Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation
Figure 2 for Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation
Figure 3 for Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation
Figure 4 for Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation
Viaarxiv icon

FLAME: A Self-Adaptive Auto-labeling System for Heterogeneous Mobile Processors

Add code
Mar 03, 2020
Figure 1 for FLAME: A Self-Adaptive Auto-labeling System for Heterogeneous Mobile Processors
Figure 2 for FLAME: A Self-Adaptive Auto-labeling System for Heterogeneous Mobile Processors
Figure 3 for FLAME: A Self-Adaptive Auto-labeling System for Heterogeneous Mobile Processors
Figure 4 for FLAME: A Self-Adaptive Auto-labeling System for Heterogeneous Mobile Processors
Viaarxiv icon