CATNIP：通过校准和令牌化的负偏好对齐实现LLM去学习
CALIBURN: Self-Calibrated LLM Unlearning Alignment
AI总结：
CATNIP通过校准和令牌化的负偏好对齐方法，实现有效LLM去学习，无需保留数据或对比对，提升知识遗忘与保留的平衡。
摘要：
预训练知识在LLM中存储引发了安全和隐私方面的重大关切，这推动了LLM去学习作为一种技术，用于选择性地移除不受欢迎的知识影响。现有方法基于梯度上升（GA），在依赖保留数据或精心挑选的对比对时往往会退化一般领域知识，这可能既不实际又数据和计算上具有约束性。负偏好对齐已被探索用于去学习以克服GA的局限性，然而，其仍然受限于参考模型的选择，并在现实数据设置中表现出削弱的性能。这些局限性提出了两个关键问题：i）我们能否实现一种有效的去学习，该方法可以量化模型对不受欢迎知识的置信度，并利用它更精确地校准梯度更新，从而减少灾难性遗忘？ii）我们能否使去学习对数据稀缺和长度变化具有鲁棒性？我们通过CATNIP（校准和令牌化的负偏好对齐）回答了这两个问题，这是一种原则性的方法，通过按模型令牌级置信度重新缩放去学习效果，从而确保对遗忘的精细控制。在MUSE和WMDP基准上的广泛评估表明，我们的工作实现了有效的去学习，而无需保留数据或对比去学习响应对，与最先进的方法相比，在知识遗忘与保留的权衡上更优。
英文摘要：
LLM unlearning aims to remove the influence of undesirable knowledge from pretrained language models, which offers a practical mechanism for addressing safety and privacy concerns. Existing unlearning approaches, such as Gradient Ascent, are prone to catastrophic forgetting. Alignment-based approaches provide an alternative direction, yet their effectiveness is limited by the quality of the reference model. In realistic settings, both methods still require large retention datasets to preserve general knowledge. We propose a principled method that quantifies the target LLM's confidence in undesirable knowledge and uses it to calibrate the model's unlearning gradient updates more precisely. It enables fine-grained control over forgetting while better preserving model utility, thus reducing the dependence on retention data or prohibitive unlearning training data. Extensive evaluations on multiple benchmarks, including MUSE and WMDP, show that our method achieves effective unlearning and improves the trade-off between knowledge removal and utility preservation compared with state-of-the-art methods.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
