研究人员开发了Divergence Decoding (DD) 方法，这是一种在推理时从大型语言模型中移除敏感数据的新颖方法。该技术采用小型辅助模型来引导LLM的输出远离特定信息，从而在不显著损失模型效用的情况下降低隐私和版权风险。在移除数据基准测试中，DD表现优于当前最先进的方法，并显示出泛化到图像生成等其他领域的潜力。 AI
影响 通过在不损害模型效用的情况下实现数据移除，为减轻LLM中的隐私和版权风险提供了一种新方法。
排序理由 该集群包含一篇详细介绍LLM移除数据新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.31293v1 Announce Type: new Abstract: Large Language Models (LLMs) frequently memorize sensitive training data thereby creating significant privacy and copyright risks. Addressing these risks, i.e., removing such knowledge from an existing model checkpoint, has proven c…
Large Language Models (LLMs) frequently memorize sensitive training data thereby creating significant privacy and copyright risks. Addressing these risks, i.e., removing such knowledge from an existing model checkpoint, has proven challenging as many unlearning methods lead to ca…