# LLM-Attacker preprint tests privacy risks in split language models

_Published Monday, October 19, 2026 at 12:17 AM EDT · Science, Security · Latest · Tier 2 — Notable_

Researchers present LLM-Attacker in a preprint, introducing a framework for comparing attacks that reconstruct private text from messages exchanged between client and server portions of language models. It records attacker access, datasets and scoring choices, saves predictions for independent scoring, and requires an undefended reference before a defence can pass.

In a controlled GPT-2 experiment, giving attackers access to gradients returned during training increased token recovery from 94.20% to 97.38% and exact-document recovery from 13.71% to 37.77%. The experiment tests how evaluation responds to changes in attacker access and is shared with a companion study. Reports distinguish implementation checks from measured attack effectiveness and identify untested attacks.

## Sources

- [cs.LG updates on arXiv.org](https://arxiv.org/abs/2610.22722)

---
Canonical: https://techandbusiness.org/newswire/WrrzZvKYh958R4OcBgBC3u
Published: 2026-10-19T04:17:07.953Z
Story chronology: 2026-10-19T04:00:00.000Z
Retrieved: 2026-10-19T06:53:07.420Z
Publisher: Tech & Business (techandbusiness.org)