mlsanityData to 14 Oct 2026

Revisiting MLLM Token Technology through the Lens of Classical Visual Coding

arXiv 2508.13460 ↗First posted 19 Aug 2025cs.CVcited by 3

On the map: Multimodal LLMs / vision-language models → Visual token pruning and efficiency

Abstract (arXiv)

Classical visual coding and Multimodal Large Language Model (MLLM) token technology share the core objective - maximizing information fidelity while minimizing computational cost. Therefore, this paper reexamines MLLM token technology, including tokenization, token compression, and token reasoning, through the established principles of long-developed visual coding area. From this perspective, we (1) establish a unified formulation bridging token technology and visual coding, enabling a systematic, module-by-module comparative analysis; (2) synthesize bidirectional insights, exploring how visual coding principles can enhance MLLM token techniques' efficiency and robustness, and conversely, how token technology paradigms can inform the design of next-generation semantic visual codecs; (3) prospect for promising future research directions and critical unsolved challenges. In summary, this study presents the first comprehensive and structured technology comparison of MLLM token and visual coding, paving the way for more efficient multimodal models and more powerful visual codecs simultaneously.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Placed in Visual token pruning and efficiency, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 45 of its 55 references that are in the map. First posted on arXiv 19 Aug 2025.

Builds on

45 of 55 references.

Closest ancestors · 9

Same direction, newest first

  1. 2025TokenCarve: Information-Preserving Visual Token Compression in Multimodal Large Language Models13 Mar
  2. 2024Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs2 Dec
  3. 2024PyramidDrop: Accelerating Your Large Vision-Language Models via Pyramid Visual Redundancy Reduction22 Oct
  4. 2024Tell Codec What Worth Compressing: Semantically Disentangled Image Coding for Machine with LMMs16 Aug
  5. 2024Bridging Compressed Image Latents and Multimodal Large Language Models29 Jul
  6. 2024High Efficiency Image Compression for Large Visual-Language Models24 Jul
  7. 2024LLaVA-PruMerge: Adaptive Token Reduction for Efficient Large Multimodal Models22 Mar
  8. 2024An Image is Worth 1/2 Tokens After Layer 2: Plug-and-Play Inference Acceleration for Large Vision-Language Models11 Mar
  9. 2023MobileVLM : A Fast, Strong and Open Vision Language Assistant for Mobile Devices28 Dec

Other references · 36

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  3. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,968
  4. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
  5. 2025Qwen2.5-VL Technical Reportcited by 5,587
  6. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  7. 2023Sigmoid Loss for Language Image Pre-Trainingcited by 3,109
  8. 2024LLaVA-OneVision: Easy Visual Task Transfercited by 2,719
  9. 2023MiniGPT-4: Enhancing Vision-Language Understanding with Advanced Large Language Modelscited by 2,711
  10. 2024How Far Are We to GPT-4V? Closing the Gap to Commercial Multimodal Models with Open-Source Suitescited by 1,050
  11. 2022Token Merging: Your ViT But Fastercited by 1,021
  12. 2018Variational image compression with a scale hyperpriorcited by 935
24 more
  1. 2025FAST: Efficient Action Tokenization for Vision-Language-Action Modelscited by 665
  2. 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 664
  3. 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 590
  4. 2020High-Fidelity Generative Image Compressioncited by 338
  5. 2022ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Codingcited by 301
  6. 2023Language Modeling Is Compressioncited by 225
  7. 2018Generative Adversarial Networks for Extreme Learned Image Compressioncited by 219
  8. 2023Learned Image Compression with Mixed Transformer-CNN Architecturescited by 211
  9. 2024AuroraCap: Efficient, Performant Video Detailed Captioning and a New Benchmarkcited by 140
  10. 2025X-Omni: Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Againcited by 112
  11. 2023DocPedia: Unleashing the Power of Large Multimodal Model in the Frequency Domain for Versatile Document Understandingcited by 106
  12. 2025Towards Practical Real-Time Neural Video Compressioncited by 80
  13. 2025HoliTom: Holistic Token Merging for Fast Video Large Language Modelscited by 58
  14. 2024Don't Look Twice: Faster Video Transformers with Run-Length Tokenizationcited by 43
  15. 2025Token Communications: A Large Model-Driven Framework for Cross-modal Context-aware Semantic Communicationscited by 35
  16. 2023TransTIC: Transferring Transformer-based Image Compression from Human Perception to Machine Perceptioncited by 32
  17. 2024MISC: Ultra-low Bitrate Image Semantic Compression Driven by Large Multimodal Modelcited by 30
  18. 2024Lossless data compression by large modelscited by 20
  19. 2024Feature Coding in the Era of Large Models: Dataset, Test Conditions, and Benchmarkcited by 12
  20. 2025Single-pass Adaptive Image Tokenization for Minimum Program Searchcited by 12
  21. 2024Rate-Distortion-Cognition Controllable Versatile Neural Image Compressioncited by 11
  22. 2025PICD: Versatile Perceptual Image Compression with Diffusion Renderingcited by 11
  23. 2022Semantic Segmentation in Learned Compressed Domaincited by 10
  24. 2025RL-RC-DoT: A Block-level RL agent for Task-Aware Video Compressioncited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 1.

3 papers.

  1. 2025StreamingTOM: Streaming Token Compression for Efficient Video Understandingcited by 43
  2. 2025OmniZip: Audio-Guided Dynamic Token Compression for Fast Omnimodal Large Language Modelscited by 22
  3. 2026Compression Tells Intelligence: Visual Coding, Visual Token Technology, and the Unificationcited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2026TokenCom: Vision-Language Model for Multimodal and Multitask Token Communications
  2. 2026Visual Token Coding for Video Multimodal Large Language Models
  3. 2024Treat Visual Tokens as Text? But Your MLLM Only Needs Fewer Efforts to See
  4. 2025Token Sequence Compression for Efficient Multimodal Computing
  5. 2025AdaTok: Adaptive Token Compression with Object-Aware Representations for Efficient Multimodal LLMs
  6. 2024From Pixels to Tokens: Byte-Pair Encoding on Quantized Visual Modalities
  7. 2024DeCo: Decoupling Token Compression from Semantic Abstraction in Multimodal Large Language Models
  8. 2024[CLS] Token Tells Everything Needed for Training-free Efficient MLLMs