Guoxuan Chen

Gausson

·

https://www.xiaohongshu.com/user/profile/615bd9080000000002018213

AI & ML interests

LLM Architecture, Pre-training, Multimodal Generation, Deep Neural Network Optimization, Sparsity

Recent Activity

authored a paper 3 days ago

SepLLM: Accelerate Large Language Models by Compressing One Segment into One Separator

authored a paper 3 days ago

Self-Adjust Softmax

authored a paper 3 days ago

Answer-Consistent Chain-of-thought Reinforcement Learning For Multi-modal Large Langauge Models

View all activity

Organizations

New activity in nvidia/kvpress-leaderboard 12 months ago

Upload the results of the training-free version of the method [SepLLM - ICML 2025 Paper](https://arxiv.org/abs/2412.12094) based on "meta-llama/Meta-Llama-3.1-8B-Instruct"

#3 opened 12 months ago by

New activity in transformers-community/support 12 months ago

Custom `generate` methods discussion

#10 opened about 1 year ago by

New activity in transformers-community/README about 1 year ago

sep_cache

#5 opened about 1 year ago by

How to integrate our method into the `transformers-community`?

#4 opened about 1 year ago by