探索个性化语言模型中的安全性-实用性权衡
Exploring Safety-Utility Trade-Offs in Personalized Language Models
AI总结：
本研究揭示了大型语言模型在个性化过程中存在的安全性-实用性权衡偏差，通过多维度评估量化该偏差，并针对开源与API模型提出基于偏好调优和提示防御的缓解策略。
摘要：
随着大型语言模型（LLMs）日益融入日常应用，确保其在不同用户群体中公平运行至关重要。本研究表明，LLMs存在个性化偏差：当模型针对用户身份进行个性化调整时，其性能会受到影响。我们通过评估LLMs在安全性和实用性两个维度的表现来量化这一偏差：安全性通过对比有无个性化时LLM对不安全提示的良性响应情况来衡量；实用性则通过评估模型在常识、数学能力、编程和推理技能等任务上的表现来衡量。研究发现，从开源模型（如Llama（Touvron等，2023）、Mistral（Jiang等，2023））到API模型（如GPT-3.5、GPT-4o（Ouyang等，2022）），均存在安全性-实用性权衡的显著差异。最后，我们探讨了基于偏好调优和提示防御的缓解策略。
英文摘要：
As large language models (LLMs) become increasingly integrated into daily applications, it is essential to ensure they operate fairly across diverse user demographics. In this work, we show that LLMs suffer from personalization bias, where their performance is impacted when they are personalized to a user's identity. We quantify personalization bias by evaluating the performance of LLMs along two axes - safety and utility. We measure safety by examining how benign LLM responses are to unsafe prompts with and without personalization. We measure utility by evaluating the LLM's performance on various tasks, including general knowledge, mathematical abilities, programming, and reasoning skills. We find that various LLMs, ranging from open-source models like Llama (Touvron et al., 2023) and Mistral (Jiang et al., 2023) to API-based ones like GPT-3.5 and GPT-4o (Ouyang et al., 2022), exhibit significant variance in performance in terms of safety-utility trade-offs depending on the user's identity. Finally, we discuss several strategies to mitigate personalization bias using preference tuning and prompt-based defenses.
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。
