Song's picture

3 5

Song

songhan

·

https://songhan.mit.edu

AI & ML interests

efficient AI computing

Recent Activity

upvoted a collection 23 days ago

upvoted a paper about 1 month ago

OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM

upvoted a paper about 2 months ago

SANA-Video: Efficient Video Generation with Block Linear Diffusion Transformer

View all activity

Organizations

authored a paper 9 months ago

LServe: Efficient Long-sequence LLM Serving with Unified Sparse Attention

Paper • 2502.14866 • Published Feb 20 • 13

authored 2 papers over 1 year ago

Wolf: Captioning Everything with a World Summarization Framework

Paper • 2407.18908 • Published Jul 26, 2024 • 32

$VILA^2$: VILA Augmented VILA

Paper • 2407.17453 • Published Jul 24, 2024 • 41

authored 2 papers almost 2 years ago

BitDelta: Your Fine-Tune May Only Be Worth One Bit

Paper • 2402.10193 • Published Feb 15, 2024 • 22

VILA: On Pre-training for Visual Language Models

Paper • 2312.07533 • Published Dec 12, 2023 • 23

authored a paper about 2 years ago

LongLoRA: Efficient Fine-tuning of Long-Context Large Language Models

Paper • 2309.12307 • Published Sep 21, 2023 • 89