Preprint
Nov 2026
The Invocation Bottleneck in LLM-Based Lossless Compression: What Can Be Removed and What Must Be Traded
This work removes execution redundancy with a GPU-resident pipeline whose block-parallel encoding is made lossless in half precision by batch-invariant kernels, speeding up arithmetic coding by up to 5.8x in encoding and 3.0x in decoding without changing its compression rate.
Zhe Yan, Yi-Hao Liu, Antonio Boffa et al.
· 0 citations