Jiajun Kang

kangjiajun

1 2

·

AI & ML interests

Efficient LLM inference, KV cache optimization, quantization, speculative decoding, model pruning

Recent Activity

liked a dataset 3 days ago

makora-ai/speculative-decoding-dataset

liked a model 3 days ago

hoho0106tw/Femh_Pruning_med4270B_awq-model

upvoted a paper 3 days ago

Complex KDA: Understanding and Enhancing the Expressivity of Kimi Delta Attention

View all activity

Organizations

None yet

models 0

None public yet

datasets 0

None public yet