Xiangpeng Wei

pemywei

https://pemywei.github.io/

pemywei

AI & ML interests

My research interests include Machine Translation and Natural Language Generation. Currently, I focus on semantic-augmentation and learning deep models for NMT as well as learning universal representations across languages.

Organizations

pemywei's activity

upvoted 3 papers 11 months ago

Cached Transformers: Improving Transformers with Differentiable Memory Cache

Paper • 2312.12742 • Published Dec 20, 2023 • 12

Weight subcloning: direct initialization of transformers using larger pretrained ones

Paper • 2312.09299 • Published Dec 14, 2023 • 17

Self-Evaluation Improves Selective Generation in Large Language Models

Paper • 2312.09300 • Published Dec 14, 2023 • 14

upvoted 2 collections 11 months ago

LLM

Collection

Multimodal LLM • 238 items • Updated Sep 26 • 10

Instruct

Collection

127 items • Updated Jul 22 • 5

upvoted 7 papers about 1 year ago

Zephyr: Direct Distillation of LM Alignment

Paper • 2310.16944 • Published Oct 25, 2023 • 121

Sorted LLaMA: Unlocking the Potential of Intermediate Layers of Large Language Models for Dynamic Inference Using Sorted Fine-Tuning (SoFT)

Paper • 2309.08968 • Published Sep 16, 2023 • 22

Optimized Network Architectures for Large Language Model Training with Billions of Parameters

Paper • 2307.12169 • Published Jul 22, 2023 • 9

Challenges and Applications of Large Language Models

Paper • 2307.10169 • Published Jul 19, 2023 • 47

upvoted a paper over 1 year ago

PolyLM: An Open Source Polyglot Large Language Model

Paper • 2307.06018 • Published Jul 12, 2023 • 25