# The Choice of 4-Bit Quantizer Decides Whether a Fine-Tuned Model Leaks Private Data

Researchers found that calibration-free GGUF Q4_K_M reproduced 5.3 percent of planted private records while calibration-based AWQ and GPTQ reproduced none, at the same bit width.

- Published: 2026-09-23T11:26:37.007Z
- Canonical: https://polylog.news/ai/2026-09-23/the-choice-of-4-bit-quantizer-decides-whether-a-fine-tuned-m
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv cs.CR](https://arxiv.org/abs/2609.25014)

Organizations routinely fine-tune small language models on private data and then compress them to four bits so they can run on modest hardware. A paper published on arXiv by researchers Cristhian Kapelinski and Diego Kreutz, presented at th…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-23/the-choice-of-4-bit-quantizer-decides-whether-a-fine-tuned-m (subscription information: https://polylog.news/pricing).