Researchers have developed StateSMix, a novel lossless compression algorithm that utilizes Mamba-style State Space Models (SSMs) combined with sparse n-gram context mixing. This system trains token-by-token on the data being compressed, eliminating the need for pre-trained weights or GPUs. StateSMix achieves competitive compression ratios, outperforming xz (LZMA2) on the enwik8 benchmark by up to 8.7%. The implementation is in pure C and can process approximately 2,000 tokens per second on standard hardware. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Introduces a new method for lossless compression using state space models, potentially improving data storage efficiency.
RANK_REASON This is a research paper detailing a new algorithm and its performance on a standard benchmark. [lever_c_demoted from research: ic=1 ai=1.0]