mlsanityData to 7 Oct 2026

Lossless data compression by large models

arXiv 2407.07723 ↗First posted 24 Jun 2024cs.IT cs.AI math.ITcited by 20

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 20 papers in the map (data through 7 Oct 2026). Builds on 24 of its 51 references that are in the map. First posted on arXiv 24 Jun 2024.

Builds on

24 of 51 references.

References · 24

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  4. 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,375
  5. 2019Common Voice: A Massively-Multilingual Speech Corpuscited by 1,358
  6. 2016End-to-end Optimized Image Compressioncited by 673
  7. 2021VoxPopuli: A Large-Scale Multilingual Speech Corpus for Representation Learning, Semi-Supervised Learning and Interpretationcited by 455
  8. 2022ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Codingcited by 296
  9. 2023Language Modeling Is Compressioncited by 225
  10. 2021Deep Contextual Video Compressioncited by 216
  11. 2022Lossy Image Compression with Conditional Diffusion Modelscited by 165
  12. 2022Hybrid Spatial-Temporal Entropy Modelling for Neural Video Compressioncited by 123
12 more
  1. 2024Neural Video Compression with Feature Modulationcited by 115
  2. 2022Pile of Law: Learning Responsible Data Filtering from the Law and a 256GB Open-Source Legal Datasetcited by 113
  3. 2019Meta-learning of Sequential Strategiescited by 68
  4. 2017Generative Compressioncited by 40
  5. 2024Lossy Image Compression with Foundation Diffusion Modelscited by 34
  6. 2024Learning Universal Predictorscited by 29
  7. 2024Idempotence and Perceptual Image Compressioncited by 27
  8. 2024Beyond Language Models: Byte Models are Digital World Simulatorscited by 21
  9. 2020MeDAL: Medical Abbreviation Disambiguation Dataset for Natural Language Understanding Pretrainingcited by 18
  10. 2021LC-FDNet: Learned Lossless Image Compression with Frequency Decomposition Networkcited by 12
  11. 2023Approximating Human-Like Few-shot Learning with GPT-based Compressioncited by 7
  12. 2023A Theory of Human-Like Few-Shot Learningcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 1; 2025: 7; 2026: 12.

20 papers.

Sort:

By area:

  1. 2024Neural Entropycited by 7
  2. 2025Information Capacity: Evaluating the Efficiency of Large Language Models via Text Compressioncited by 7
  3. 2025Mechanistic Indicators of Understanding in Large Language Modelscited by 6
  4. 2025Revisiting MLLM Token Technology through the Lens of Classical Visual Codingcited by 3
  5. 2025SelfAI: A self-directed framework for long-horizon scientific discoverycited by 2
  6. 2026Compression Tells Intelligence: Visual Coding, Visual Token Technology, and the Unificationcited by 2
  7. 2026AgentGC: Evolutionary Learning-based Lossless Compression for Genomics Data with LLM-driven Multiple Agentcited by 1
  8. 2026Grokking From Abstraction to Intelligencecited by 1
  9. 2026Adapting Diffusion Language Models for Lossless Pixel-Level Image Transmissioncited by 1
  10. 2025Way to Build Native AI-driven 6G Air Interface: Principles, Roadmap, and Outlookcited by 0
  11. 2025The Algorithmic Regulatorcited by 0
  12. 2025Shakespeare, Entropy and Educated Monkeyscited by 0
8 more
  1. 2026Benchmarking Language Modeling for Lossless Compression of Full-Fidelity Audiocited by 0
  2. 2026NeuralLVC: Neural Lossless Video Compression via Masked Diffusion with Temporal Conditioningcited by 0
  3. 2026Thinking as Compression: Your Reasoning Model is Secretly a Context Compressorcited by 0
  4. 2026Text Distance from Nested and Hierarchical Repetitions: A Compression-Based Perspectivecited by 0
  5. 2026LUMI: Tokenizer-Agnostic LLM-Based Lossless Image Compressioncited by 0
  6. 2026Beyond Representational Similarity: Source-Conditioned Description-Length Gain for Generative Plagiarism Detection and Candidate Source Rerankingcited by 0
  7. 2026FLM: Frequency-Aware Language Models for Generative Image Compressioncited by 0
  8. 2026HyperZip: Efficient Data Compression through Personalized Diffusion LLMs with Hypernetworkscited by 0