UNITES Lab

university

https://tianlong-chen.github.io/

unites-lab

AI & ML interests

AI Trustworthiness, Efficiency, and for Science

Recent Activity

pingzhili authored a paper 7 days ago

Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild

pingzhili authored a paper 7 days ago

Glider: Global and Local Instruction-Driven Expert Router

pingzhili authored a paper 7 days ago

PortLLM: Personalizing Evolving Large Language Models with Training-Free and Portable Model Patches

View all activity

unites-lab's activity

pingzhili

authored 3 papers 7 days ago

Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild

Paper • 2410.05357 • Published Oct 7, 2024

Glider: Global and Local Instruction-Driven Expert Router

Paper • 2410.07172 • Published Oct 9, 2024

PortLLM: Personalizing Evolving Large Language Models with Training-Free and Portable Model Patches

Paper • 2410.10870 • Published Oct 8, 2024 • 1

pingzhili

published a model 3 months ago

unites-lab/pythia-70m-moepp-6e-20k

pingzhili

updated a model 3 months ago

unites-lab/pythia-70m-moepp-6e-20k

pingzhili

updated a collection 5 months ago

(Internal) MoE Expert Collaboration

0 items • Updated Dec 5, 2024

pingzhili

authored 2 papers about 1 year ago

Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy

Paper • 2310.01334 • Published Oct 2, 2023 • 3

Revisiting Zeroth-Order Optimization for Memory-Efficient LLM Fine-Tuning: A Benchmark

Paper • 2402.11592 • Published Feb 18, 2024 • 2