A tokenizer-free pixel embedding, position encoding, and 256-way head let frozen LLMs act as portable entropy models for lossless RGB compression across model families.
Lossless data compression by large models , volume =
2 Pith papers cite this work, alongside 8 external citations. Polarity classification is still indexing.
2
Pith papers citing it
8
external citations · OpenAlex
years
2026 2representative citing papers
LLMs exhibit Bayesian-like hypothesis updating with strong-sampling bias and an evaluation-generation gap but generalize poorly outside observed data.
citing papers explorer
-
LUMI: Tokenizer-Agnostic LLM-Based Lossless Image Compression
A tokenizer-free pixel embedding, position encoding, and 256-way head let frozen LLMs act as portable entropy models for lossless RGB compression across model families.
-
Hypothesis generation and updating in large language models
LLMs exhibit Bayesian-like hypothesis updating with strong-sampling bias and an evaluation-generation gap but generalize poorly outside observed data.