LectureRechercheRadarCadre d'investissement
Connexion / Inscription
← Retour à la lecture
ActualitésSujet · LLMONMatériel2 min de lectureRédaction DisconfirmAI

IBM Releases LLMON Markup Language and Experimental Granite Preview Models to Combat Prompt Injection

IBM releases LLMON, a new markup language using special tokens to separate instructions from data, along with two experimental Granite 4.1 preview models trained to use it for improved prompt injection resistance.

2026-11-13 03:07 GMT+8

IBM has released LLMON (LLM Object Notation), a new markup language designed specifically for large language models, along with two experimental preview models: granite-4.1-3b-LLMON-preview and granite-4.1-8b-LLMON-preview. These models are derived from the Granite 4.1 base checkpoints and are trained to consume and generate text marked with LLMON tags.

The core innovation of LLMON is its use of reserved special tokens to explicitly mark structural components like instructions, data, and requirements. Unlike JSON, which relies on characters that can appear in normal text, LLMON uses unique delimiters that allow the model to distinguish between executable commands and passive data payloads. This structural separation aims to mitigate indirect prompt injection attacks, where malicious instructions are hidden within retrieved documents or user data.

According to IBM's internal evaluation on the OPI benchmark, the LLMON models demonstrated significantly higher robustness against static indirect prompt injections compared to other open-source and closed-source models. The company reports that these models avoided 47% more attacks than the best frontier model tested and 95% more than the best open-source model, while maintaining general capability comparable to the standard Granite 4.1 Instruct models.

The release includes a GitHub repository with conversion tools that allow users to translate existing JSON pipelines into LLMON format losslessly. Additionally, IBM provides notebooks demonstrating how to fine-tune LoRA adapters on top of the LLMON models to understand custom domain-specific vocabularies. These are experimental previews intended for research and evaluation, not for production deployment.

Connexion / Inscription
LectureRechercheRadarCadre d'investissement

Sources

  • huggingface.co↗
Sujet

LLMON

IBM于2026年11月12日发布LLMON(LLM Object Notation),这是一种专为大语言模型设计的原生标记语言,旨在通过架构层面的结构分离来防御间接提示注入攻击。LLMON利用保留的特殊令牌明确区分“指令”与“数据”,并引…

4 étapes1 piècesMis à jour 2026-11-17 21:27

Comment on en est arrivé là

  1. 2026-10-15IBM Research团队在arXiv提交题为《LLMON: An LLM-native Markup Language to Leverage Structure and Semantics at the LLM Interface》的论文,详细阐述了LLMON的设计需求、语法结构及其在提升模型准确性、安全性和可解释性方面的初步实证依据。arxiv.org ↗
  2. 2026-10-28IBM团队提交第二篇论文《Post-Training Large Language Models with LLMON》,介绍用于构建LLMON预览模型的后训练方法论,包括4.6B token的结构化语料库构建及注意力层级的可解释性分析。arxiv.org ↗
  3. 2026-11-12IBM Granite团队在Hugging Face博客正式发布LLMON标记语言,并推出两款实验性预览模型granite-4.1-3b-LLMON-preview和granite-4.1-8b-LLMON-preview,同步开源包含转换器和工作流示例的GitHub仓库。huggingface.co ↗

Ce qui reste non résolu

  • LLMON在更广泛的非IBM自研基准或第三方红队测试中的实际防御效果如何?
  • 对于已深度依赖JSON协议的企业级应用,无损转换工具在实际生产环境中的性能开销和兼容性表现怎样?
  • 除了Granite系列,其他主流开源或闭源模型是否计划支持LLMON或类似的指令-数据分离标记标准?
Lire le sujet complet →
Continuer la recherche →