Metraipolis › archive › dc_62fa05e17aa3
Security-Enhanced Seed-Based Weight Quantization for Large Language Models — arXiv:2609.38477
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38477
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T12:43:47.570Z
- SHA-256 of the archived text
fde299243c1043e0c5e94e4418c164fc146961a3b0aca69b73a6f172ed3daf2a- Stable URL of this page
https://metraipolis.com/documento/dc_62fa05e17aa3
How to cite
Security-Enhanced Seed-Based Weight Quantization for Large Language Models — arXiv:2609.38477 — arXiv (en), https://arxiv.org/abs/2609.38477 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_62fa05e17aa3 (Metraipolis archive dc_62fa05e17aa3, sha256 fde299243c1043e0c5e94e4418c164fc146961a3b0aca69b73a6f172ed3daf2a)
BibTeX
@misc{dc62fa05e17aa3,
title = {Security-Enhanced Seed-Based Weight Quantization for Large Language Models — arXiv:2609.38477},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38477},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_62fa05e17aa3; sha256 fde299243c1043e0c5e94e4418c164fc146961a3b0aca69b73a6f172ed3daf2a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38477. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) incur substantial storage, memory-bandwidth and energy costs, motivating compact weight representations. Existing seed-based compression methods reconstruct weights from compact pseudo-random representations but do not explicitly account for the non-uniform sensitivity of model weights. We introduce Seed-Q, a security-enhanced sensitivity-aware seed-based weight compression framework that uses lightweight Linear Feedback Shift Register (LFSR)-based weight generation with non-uniform bit allocation. Our approach assigns larger representation budgets to sensitive weights while aggressively compressing less sensitive regions. Importantly, this non-uniform allocation requires no side-information: the decoder deterministically reconstructs the bit-allocation schedule, with no rung depending on the decoded weights, eliminating the need to store per-block metadata or use calibration data while preserving the baseline coding rate. Experiments across diverse LLMs show that Seed-Q matches 4-bit perplexity of SeedLM with fewer bits, while at the same 4 bits/weight it reduces both perplexity degradation and zero-shot accuracy loss relative to SeedLM. We also show that Seed-Q simultaneously achieves high security against bit-flip attacks on model parameters, as bit corruption affects multiple reconstructed weights, greatly amplifying its impact and making it easier to detect. We further implement Seed-Q in an ASIC-based accelerator and demonstrate modest hardware overhead compared to prior seed-based approaches. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-01 in cs.AI). Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.38477 — 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 2026-10-07T12:14:13.886Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).