HyperZip: Efficient Data Compression through Personalized Diffusion LLMs with Hypernetworks
This work proposes HyperZip, an efficient and scalable LLM-based compression framework that leverages diffusion-based LLMs (dLLMs) with Multi-Token Prediction (MTP) to accelerate LLM-based data compression processes and identifies a trade-off in diffusion-based compression.
T. Nguyen, Khang Tran, Nhathai Phan
· 0 citations