arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.12934cs.CRcs.AIcs.HC
收藏

大语言模型风险与AI护栏的现状

Current state of LLM Risks and AI Guardrails

Carnegie Mellon University(卡内基梅隆大学)

Suriya Ganesh Ayyamperumal, Limin Ge

AI总结:

本研究梳理大语言模型的各类固有风险,评估现有护栏与模型对齐技术,提出分层防护等安全策略,指出平衡多维度需求的挑战,强调持续研发对大模型安全落地的重要性。

摘要:

大语言模型(LLMs)已变得日益复杂,被广泛部署于安全与可靠性至关重要的敏感应用中。然而,LLMs 自身存在固有风险,包括偏见、引发不安全行为的可能性、数据集投毒、可解释性缺失、幻觉以及不可复现性。这些风险使得开发“护栏”以让 LLMs 与预期行为对齐、减轻潜在危害成为必要。\n 本研究探讨了部署 LLMs 相关的风险,并评估了当前实现护栏与模型对齐技术的各类方法。我们研究了内在与外在偏见评估方法,讨论了公平性指标对负责任 AI 开发的重要性。本文探究了智能体式 LLMs(能够执行现实世界行动的 LLM)的安全性与可靠性,强调了可测试性、故障安全机制以及态势感知的必要性。\n 本文提出了保障 LLMs 安全的技术策略,包括在外部、次级和内部层级运行的分层防护模型。重点介绍了系统提示词、检索增强生成(RAG)架构,以及减少偏见、保护隐私的相关技术。\n 有效的护栏设计需要深入理解 LLM 的预期用例、相关法规以及伦理考量。在准确性与隐私等相互冲突的需求之间取得平衡,仍是一项持续存在的挑战。本研究强调了持续开展研发工作的重要性,以确保 LLMs 在现实应用中得到安全、负责任的使用。

英文摘要:

Large language models (LLMs) have become increasingly sophisticated, leading to widespread deployment in sensitive applications where safety and reliability are paramount. However, LLMs have inherent risks accompanying them, including bias, potential for unsafe actions, dataset poisoning, lack of explainability, hallucinations, and non-reproducibility. These risks necessitate the development of "guardrails" to align LLMs with desired behaviors and mitigate potential harm. This work explores the risks associated with deploying LLMs and evaluates current approaches to implementing guardrails and model alignment techniques. We examine intrinsic and extrinsic bias evaluation methods and discuss the importance of fairness metrics for responsible AI development. The safety and reliability of agentic LLMs (those capable of real-world actions) are explored, emphasizing the need for testability, fail-safes, and situational awareness. Technical strategies for securing LLMs are presented, including a layered protection model operating at external, secondary, and internal levels. System prompts, Retrieval-Augmented Generation (RAG) architectures, and techniques to minimize bias and protect privacy are highlighted. Effective guardrail design requires a deep understanding of the LLM's intended use case, relevant regulations, and ethical considerations. Striking a balance between competing requirements, such as accuracy and privacy, remains an ongoing challenge. This work underscores the importance of continuous research and development to ensure the safe and responsible use of LLMs in real-world applications.

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。

↑