Picture for Shuaiwen Leon Song

Shuaiwen Leon Song

CorDA: Context-Oriented Decomposition Adaptation of Large Language Models

Add code
Jun 07, 2024
Viaarxiv icon

Dragonfly: Multi-Resolution Zoom Supercharges Large Visual-Language Model

Add code
Jun 03, 2024
Viaarxiv icon

FP6-LLM: Efficiently Serving Large Language Models Through FP6-Centric Algorithm-System Co-Design

Add code
Jan 25, 2024
Viaarxiv icon

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

Add code
Oct 11, 2023
Viaarxiv icon

Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity

Add code
Sep 19, 2023
Viaarxiv icon

RenAIssance: A Survey into AI Text-to-Image Generation in the Era of Large Model

Add code
Sep 02, 2023
Viaarxiv icon

DeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scales

Add code
Aug 02, 2023
Viaarxiv icon

HEAT: A Highly Efficient and Affordable Training System for Collaborative Filtering Based Recommendation on CPUs

Add code
May 03, 2023
Viaarxiv icon

Brief Industry Paper: The Necessity of Adaptive Data Fusion in Infrastructure-Augmented Autonomous Driving System

Add code
Jul 02, 2022
Figure 1 for Brief Industry Paper: The Necessity of Adaptive Data Fusion in Infrastructure-Augmented Autonomous Driving System
Figure 2 for Brief Industry Paper: The Necessity of Adaptive Data Fusion in Infrastructure-Augmented Autonomous Driving System
Figure 3 for Brief Industry Paper: The Necessity of Adaptive Data Fusion in Infrastructure-Augmented Autonomous Driving System
Figure 4 for Brief Industry Paper: The Necessity of Adaptive Data Fusion in Infrastructure-Augmented Autonomous Driving System
Viaarxiv icon

COMET: A Novel Memory-Efficient Deep Learning Training Framework by Using Error-Bounded Lossy Compression

Add code
Nov 18, 2021
Figure 1 for COMET: A Novel Memory-Efficient Deep Learning Training Framework by Using Error-Bounded Lossy Compression
Figure 2 for COMET: A Novel Memory-Efficient Deep Learning Training Framework by Using Error-Bounded Lossy Compression
Figure 3 for COMET: A Novel Memory-Efficient Deep Learning Training Framework by Using Error-Bounded Lossy Compression
Figure 4 for COMET: A Novel Memory-Efficient Deep Learning Training Framework by Using Error-Bounded Lossy Compression
Viaarxiv icon