---
id: 20260317-T0-23
title: "GONE方法实现大模型结构化知识遗忘"
title_en: "GONE Method Enables Structured Knowledge Unlearning in LLMs"
url: https://ai.daily.yangsir.net/daily/20260317-T0-23
issue_date: 2026-03-17
publish_date: 2026-03-16T04:00:00.000Z
source_name: "arXiv cs.CL (NLP)"
source_url: https://arxiv.org/abs/2603.12275
---

# GONE方法实现大模型结构化知识遗忘

arXiv论文2603.12275提出GONE方法，通过邻域扩展分布实现大语言模型（LLM）的结构化知识遗忘。现有知识遗忘方法常受限于高维噪声和模型结构约束。GONE通过分析知识在模型参数中的分布，精确定位需要遗忘的内容，同时保留其他知识。实验显示，该方法在移除特定事实后，模型准确率下降仅5%，而其他知识保持率超95%，显著优于现有方法。

## English Version

**GONE Method Enables Structured Knowledge Unlearning in LLMs**

arXiv paper 2603.12275 introduces GONE, a method for structured knowledge unlearning in Large Language Models (LLMs). Existing approaches struggle with high-dimensional noise and structural constraints. GONE precisely targets unwanted knowledge while preserving other information by analyzing parameter distributions. Tests show it reduces target fact accuracy by just 5% while maintaining 95%+ retention of other knowledge, outperforming previous methods.

---

**来源**：[arXiv cs.CL (NLP)](https://arxiv.org/abs/2603.12275)

**详情页**：https://ai.daily.yangsir.net/daily/20260317-T0-23

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*