Kwanhee Lee

kwanhee

·

https://kwanhee-lee.github.io

AI & ML interests

Model compression, optimization

Recent Activity

upvoted a paper about 18 hours ago

FLAMIA: FLAsh MIcroscaling Attention for Fast and Causally Safe Training on Blackwell

new activity about 21 hours ago

ISTA-DASLab/Kimi-K3-P48NVFP4-MoESQ:Kimi K3 GSQ-RCO-GGUF IQ1_S, IQ1_M, IQ2_XXS

new activity about 22 hours ago

ISTA-DASLab/Kimi-K3-P48NVFP4-MoESQ:12 indexed weight shards missing / 404 EntryNotFound (revision 57ee129)

View all activity

Organizations

upvoted a paper about 18 hours ago

FLAMIA: FLAsh MIcroscaling Attention for Fast and Causally Safe Training on Blackwell

Paper • 2611.03344 • Published 1 day ago • 17

upvoted a paper 14 days ago

Hardware-Native Joint Sparse-Quantization for Trillion-Scale Mixture-of-Experts

Paper • 2610.02241 • Published Sep 29 • 7

upvoted a collection 29 days ago

MoESQ

Paired-4:8 + NVFP4 W4A4 expert compressed MoE models for NVIDIA Blackwell Sparse Tensor Cores. • 6 items • Updated 13 days ago • 6