Metraipolis › archive › dc_f27abac0b85b
Compressing AI Traffic: Standardized Neural Network Coding of Visual-Token Representations in Split Vision-Language Inference — arXiv:2609.01200
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.01200
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:13:50.138Z
- SHA-256 of the archived text
8c29659f333393ff97b62244e2d2f49063ff37a80be270913037fbfc7a14a13d- Stable URL of this page
https://metraipolis.com/documento/dc_f27abac0b85b
How to cite
Compressing AI Traffic: Standardized Neural Network Coding of Visual-Token Representations in Split Vision-Language Inference — arXiv:2609.01200 — arXiv (en), https://arxiv.org/abs/2609.01200 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f27abac0b85b (Metraipolis archive dc_f27abac0b85b, sha256 8c29659f333393ff97b62244e2d2f49063ff37a80be270913037fbfc7a14a13d)
BibTeX
@misc{dcf27abac0b85b,
title = {Compressing AI Traffic: Standardized Neural Network Coding of Visual-Token Representations in Split Vision-Language Inference — arXiv:2609.01200},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.01200},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f27abac0b85b; sha256 8c29659f333393ff97b62244e2d2f49063ff37a80be270913037fbfc7a14a13d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01200. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
When the visual encoder and the language decoder of a vision-language model (VLM) run on different compute nodes, the intermediate visual-token embeddings become a communicated payload rather than an internal activation. We call such machine-consumed intermediate tensors AI traffic and ask how far they can be compressed with a standardized, training-free codec. We insert ISO/IEC 15938-17 Neural Network Coding (NNC) round trips on the complete visual interface of a Qwen3-VL-8B-Instruct video question answering pipeline, comprising the main visual-token representation and the DeepStack feature streams, while leaving weights, prompts, and generation untouched, and sweep the quantization parameter (QP) over a wide rate range. Closed-ended Video-MME accuracy remains close to the uncompressed reference up to a 98% reduction of the transmitted BF16 tensor and only then collapses; open-ended MLVU generation shows the same plateau-and-collapse profile under an LLM judge. This robustness is not due to near-lossless reconstruction: the decoded tensor is heavily discretized, carries substantial row-wise relative L2 error, and has a visibly steeper singular-value decay than its source. Downstream reasoning therefore depends on coarse structure and relative geometry rather than exact floating-point values, which argues for rate-task rather than rate-distortion optimization of AI traffic codecs. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-02 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Image and Video Processing (eess.IV). Pagina del lavoro: https://arxiv.org/abs/2609.01200 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T07:00:58.971Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).