Metraipolis › archive › dc_b7eb10d186cb

BARQ: Balanced Codebook Refinement for Low-Bit LLM Quantization — arXiv:2610.04490

Source
arXiv
Original
https://arxiv.org/abs/2610.04490
License (as archived)
CC0-1.0
Archived
2026-10-06T05:30:35.431Z
SHA-256 of the archived text
ef62fbf25fec3bdafbeefbfd80af88cdf198ffb90d250e14b0866f13e5b76906
Stable URL of this page
https://metraipolis.com/documento/dc_b7eb10d186cb

How to cite

BARQ: Balanced Codebook Refinement for Low-Bit LLM Quantization — arXiv:2610.04490 — arXiv, https://arxiv.org/abs/2610.04490 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b7eb10d186cb (Metraipolis archive dc_b7eb10d186cb, sha256 ef62fbf25fec3bdafbeefbfd80af88cdf198ffb90d250e14b0866f13e5b76906)
BibTeX
@misc{dcb7eb10d186cb,
  title = {BARQ: Balanced Codebook Refinement for Low-Bit LLM Quantization — arXiv:2610.04490},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04490},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b7eb10d186cb; sha256 ef62fbf25fec3bdafbeefbfd80af88cdf198ffb90d250e14b0866f13e5b76906},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04490. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

As large language models (LLMs) grow in parameter count, model storage and parameter memory traffic have become major bottlenecks to efficient deployment. Codebook-based weight quantization reduces these costs, but imbalanced nearest-codeword assignments during fitting can leave some codewords insufficiently updated, limiting effective codebook utilization. To address this limitation, we propose Balanced Assignment Refinement for Quantization (BARQ), which improves quantization quality through balanced fitting of existing codebooks. Specifically, we first compute joint soft assignments between weight blocks and codewords through entropically regularized optimal transport with uniform marginals and curvature-weighted reconstruction costs, ensuring equal positive fitting mass for every codeword in the exact solution. We then refine the codewords through an assignment-weighted barycentric update, which we prove minimizes the fitting objective for fixed assignments. For finite Sinkhorn iterations, the implemented update retains this optimality provided all codeword masses exceed the denominator floor. Finally, we discard the soft assignments and use the refined codebook for standard hard nearest-codeword encoding, with our analysis establishing sufficient conditions for reducing hard-quantization distortion and evaluation loss. Across multiple LLMs, BARQ achieves lower perplexity and higher mean zero-shot accuracy than the evaluated baselines at comparable bit budgets. The code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.04490 — 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 2026-10-06T05:27:05.685Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).