Metraipolis › archive › dc_9ef720d1470b
Quantize by Drift: Label-Free Mixed-Precision Post-Training Quantization for Text Embedders — arXiv:2610.09227
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.09227
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T11:26:30.646Z
- SHA-256 of the archived text
ed687f2f52fc61229561bf35a6cc4203b6ab7ab5de2e13c81aad03323a5ac390- Stable URL of this page
https://metraipolis.com/documento/dc_9ef720d1470b
How to cite
Quantize by Drift: Label-Free Mixed-Precision Post-Training Quantization for Text Embedders — arXiv:2610.09227 — arXiv (en), https://arxiv.org/abs/2610.09227 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9ef720d1470b (Metraipolis archive dc_9ef720d1470b, sha256 ed687f2f52fc61229561bf35a6cc4203b6ab7ab5de2e13c81aad03323a5ac390)
BibTeX
@misc{dc9ef720d1470b,
title = {Quantize by Drift: Label-Free Mixed-Precision Post-Training Quantization for Text Embedders — arXiv:2610.09227},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.09227},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9ef720d1470b; sha256 ed687f2f52fc61229561bf35a6cc4203b6ab7ab5de2e13c81aad03323a5ac390},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09227. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Mixed-precision post-training quantization needs a per-module sensitivity signal; for a text embedder the obvious one -- the retrieval quality a module costs when quantized -- needs relevance labels that deployments rarely have. We measure a label-free substitute: quantization-induced representation drift, obtained by quantizing one module, re-encoding the corpus, and recording how far the output embeddings moved from their full-precision positions. What is specific is the observable: the deployed output representation a dense retriever ranks with. Across five development embedders, configuration-level drift orders sampled mixed-precision plans against held-out retrieval quality at a macro Spearman of 0.911, the sensitivity transports across calibration corpora and retrieval domains in the usable regime, module drifts compose rank-consistently but not numerically, and relevance-derived sensitivity adds no consistent value. The method is one additive allocation under a hard packed-byte budget, with no labels and no search. On three embedders held untouched until method, baselines and hypotheses were frozen and sealed, the pre-registered directional hypothesis against the prior LieQ criterion holds (3/3 at the main budget, no collapse) and drift scores above a two-sided LieQ steelman in 2/3; but at the main budget drift is numerically lower than same-budget uniform precision on all three (-0.99, -0.85, -1.01 points), having reduced module and whole-model drift as designed. Output drift is thus a robust coarse sensitivity signal, not a universally optimal allocation objective: it avoids the catastrophic failures of the transferred signed-geometry adaptation and can remain usable at stressed budgets where uniform collapses, but fine-grained redistribution around a strong uniform operating point remains unresolved. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Information Retrieval (cs.IR) ; Computation and Language (cs.CL); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.09227 — 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 2026-10-08T10:24:06.370Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).