Metraipolis › archive › dc_950319c4ba73

Fractional Power-of-Two Quantization for Efficient and Effective Multiplier-Free LLM Inference — arXiv:2610.21257

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.21257
License (as archived)
CC0-1.0
Archived
2026-10-16T09:12:04.033Z
SHA-256 of the archived text
bfc7801990071c03e931c68fd9861cbc3c47bf078c3a060b5af0ce18ed2971e9
Stable URL of this page
https://metraipolis.com/documento/dc_950319c4ba73

How to cite

Fractional Power-of-Two Quantization for Efficient and Effective Multiplier-Free LLM Inference — arXiv:2610.21257 — arXiv (en), https://arxiv.org/abs/2610.21257 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_950319c4ba73 (Metraipolis archive dc_950319c4ba73, sha256 bfc7801990071c03e931c68fd9861cbc3c47bf078c3a060b5af0ce18ed2971e9)
BibTeX
@misc{dc950319c4ba73,
  title = {Fractional Power-of-Two Quantization for Efficient and Effective Multiplier-Free LLM Inference — arXiv:2610.21257},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.21257},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_950319c4ba73; sha256 bfc7801990071c03e931c68fd9861cbc3c47bf078c3a060b5af0ce18ed2971e9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.21257. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Language Models (LLMs) demand efficient inference, where the multiply-accumulate (MAC) operations in linear projections dominate compute and energy cost. Post-training quantization (PTQ) reduces this cost by mapping weights and activations to low-bit integers. Power-of-Two (PoT) weight quantization promises efficient multiplier-free LLM inference via shift-add logic, but suffers significant accuracy degradation at low bitwidths due to its coarse base-2 quantization grid. We introduce Fractional PoT (FPoT), a base-$\sqrt{2}$ (half-octave) fractional PoT grid that achieves practical 4-bit weight quantization while preserving a fully integer shift-add datapath. The method combines (i) the base-$\sqrt{2}$ weight grid with uniform integer activations, (ii) a calibration-time grid-alignment refinement absorbed into the static per-channel scale, and (iii) a lightweight 1-adder $\sqrt{2}$ approximation whose error is absorbed during weight calibration via approximate-grid quantization, yielding a processing element (PE) with significantly fewer gates than a 4-bit integer multiplier. On Llama-3-8B with 4-bit weight and activation quantization, FPoT improves zero-shot commonsense reasoning performance by 8.72% relative to the PoT baseline, while maintaining performance close to the uniform integer baseline, with only a negligible 0.4%p degradation. These results hold across various configurations spanning seven models and three bit-widths. Furthermore, FPoT can be integrated with state-of-the-art LLM PTQ methods, demonstrating framework-agnostic applicability.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.21257 — 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 2026-10-16T08:44:17.927Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).