精准、低成本且灵活：通过单一向量消融缓解语言模型的误拒绝
Surgical, Cheap, and Flexible: Mitigating False Refusal in Language Models via Single Vector Ablation
AI总结：
本文提出一种无需训练且与模型无关的单一向量消融方法，通过提取并消融误拒绝向量，有效降低语言模型的误拒绝率，同时保持其安全性和通用能力，并支持细粒度的安全校准。
摘要：
训练一个既有帮助又无害的语言模型需要仔细校准拒绝行为：模型应拒绝遵循恶意指令或提供有害建议（例如“我该怎么杀某人？”），但不应拒绝安全请求，即使它们表面上类似于不安全请求（例如“我该怎么终止一个Python进程？”）。正如先前工作所示，避免这种误拒绝即使对于高能力语言模型也具挑战性。本文提出一种简单且精准的方法，通过单一向量消融缓解语言模型中的误拒绝。对于给定模型，我们提取一个误拒绝向量，并表明消融该向量能降低误拒绝率，同时保持模型安全性和通用能力。我们还表明该方法可用于模型安全的细粒度校准。我们的方法无需训练且与模型无关，使其适用于缓解当前和未来语言模型中的误拒绝问题。
英文摘要：
Training a language model to be both helpful and harmless requires careful calibration of refusal behaviours: Models should refuse to follow malicious instructions or give harmful advice (e.g."how do I kill someone?"), but they should not refuse safe requests, even if they superficially resemble unsafe ones (e.g. "how do I kill a Python process?"). Avoiding such false refusal, as prior work has shown, is challenging even for highly-capable language models. In this paper, we propose a simple and surgical method for mitigating false refusal in language models via single vector ablation. For a given model, we extract a false refusal vector and show that ablating this vector reduces false refusal rate while preserving the model's safety and general capabilities. We also show that our approach can be used for fine-grained calibration of model safety. Our approach is training-free and model-agnostic, making it useful for mitigating the problem of false refusal in current and future language models.
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。
