Pinned Repositories
ChinChunMei-LLM
Llama3-FastInference
This is a Llama3 inference project based on vLLM server and async client. This project provides at least 6 times inference speed boost compared to the huggingface inference method.
MaskLoss
A preprint for MaskLoss A Regularizing Loss by Masking Similar Labels
MTLM
A preprint for "MTLM An Innovative Language Model Training Paradigm for ASR"
RicardoLeeV587's Repositories
RicardoLeeV587/ChinChunMei-LLM
RicardoLeeV587/Llama3-FastInference
This is a Llama3 inference project based on vLLM server and async client. This project provides at least 6 times inference speed boost compared to the huggingface inference method.
RicardoLeeV587/MaskLoss
A preprint for MaskLoss A Regularizing Loss by Masking Similar Labels
RicardoLeeV587/MTLM
A preprint for "MTLM An Innovative Language Model Training Paradigm for ASR"