Lossless data compression by large models
This addresses the need for revolutionary compression ideas to meet 6G communication speed requirements, representing a new paradigm rather than an incremental improvement.
The paper tackles the problem of improving lossless data compression by leveraging large language models (LLMs), resulting in LMCompress, which doubles compression ratios for images, audio, and videos and quadruples it for texts compared to existing methods.
Modern data compression methods are slowly reaching their limits after 80 years of research, millions of papers, and wide range of applications. Yet, the extravagant 6G communication speed requirement raises a major open question for revolutionary new ideas of data compression. We have previously shown all understanding or learning are compression, under reasonable assumptions. Large language models (LLMs) understand data better than ever before. Can they help us to compress data? The LLMs may be seen to approximate the uncomputable Solomonoff induction. Therefore, under this new uncomputable paradigm, we present LMCompress. LMCompress shatters all previous lossless compression algorithms, doubling the lossless compression ratios of JPEG-XL for images, FLAC for audios, and H.264 for videos, and quadrupling the compression ratio of bz2 for texts. The better a large model understands the data, the better LMCompress compresses.