Metraipolis › archive › dc_6fb10afb0fee

E8-QAT: Efficient 2-Bit Quantization-Aware Training with Implicit Lattice Codebooks — arXiv:2610.27699

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.27699
License (as archived)
CC0-1.0
Archived
2026-10-21T07:06:46.917Z
SHA-256 of the archived text
941f5d366f2f29fa9fef56a637b2d0e8aefeffced9efd4669996a8f24b62a8aa
Stable URL of this page
https://metraipolis.com/documento/dc_6fb10afb0fee

How to cite

E8-QAT: Efficient 2-Bit Quantization-Aware Training with Implicit Lattice Codebooks — arXiv:2610.27699 — arXiv (en), https://arxiv.org/abs/2610.27699 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6fb10afb0fee (Metraipolis archive dc_6fb10afb0fee, sha256 941f5d366f2f29fa9fef56a637b2d0e8aefeffced9efd4669996a8f24b62a8aa)
BibTeX
@misc{dc6fb10afb0fee,
  title = {E8-QAT: Efficient 2-Bit Quantization-Aware Training with Implicit Lattice Codebooks — arXiv:2610.27699},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.27699},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6fb10afb0fee; sha256 941f5d366f2f29fa9fef56a637b2d0e8aefeffced9efd4669996a8f24b62a8aa},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27699. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Low-bit quantization reduces the memory footprint of large language models, but preserving accuracy with efficient training remains challenging. Scalar quantization encodes weights independently, while explicit-codebook vector quantization introduces costly search during training. We propose E8-QAT, a 2-bit quantization-aware training framework that uses the $E_8$ lattice as an implicit eight-dimensional codebook. It jointly quantizes weight groups through arithmetic projection without storing or searching an explicit codebook during the training forward pass. A 2-bit core encoding and sparse extension preserve the selected lattice points exactly, adding less than 0.1\% to core index storage. On Qwen3.6-35B-A3B, E8-QAT matches the performance of the full-precision model in terms of the average score across the four tasks. E8-QAT improves over the state-of-the-art baseline by 2.03\% on IFEval and by 1.51\% on LiveCodeBench, under the same data, schedule, and bit width. The 2-bit representation reduces peak memory by 80\% and supports a $4\times$ larger batch. At each representation's largest feasible batch, E8-QAT reaches $1.97\times$ the decoding throughput of BF16.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.27699 — 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 2026-10-21T07:05:43.437Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).