弥合安全鸿沟：用于可信大语言模型推理的护栏流水线
Bridging the Safety Gap: A Guardrail Pipeline for Trustworthy LLM Inferences
AI总结：
针对LLM推理的安全风险，提出集成安全检测器、接地、定制器和修复器的Wildflare GuardRail护栏流水线，其不安全内容检测性能可比肩OpenAI API，恶意URL处理准确率100%，幻觉修复准确率80.7%。
摘要：
我们提出Wildflare GuardRail，这是一种护栏流水线，旨在通过系统解决整个处理工作流中的风险，增强大语言模型（LLM）推理的安全性和可靠性。Wildflare GuardRail集成了多个核心功能模块，包括安全检测器（识别不安全输入、检测模型输出中的幻觉并生成根本原因解释）、接地模块（利用从向量数据库检索的信息将用户查询语境化）、定制器（使用轻量级规则包装器实时调整输出），以及修复器（利用安全检测器提供的幻觉解释纠正LLM的错误输出）。结果显示，尽管我们的安全检测器中的不安全内容检测模型是在由多个公共数据集构建的小型数据集上训练的，但其性能与OpenAI API相当；同时，轻量级包装器无需昂贵的模型调用，就能以每查询1.06秒的速度和100%的准确率处理模型输出中的恶意URL；此外，幻觉修复模型在减少幻觉方面表现出有效性，准确率达80.7%。
英文摘要：
We present Wildflare GuardRail, a guardrail pipeline designed to enhance the safety and reliability of Large Language Model (LLM) inferences by systematically addressing risks across the entire processing workflow. Wildflare GuardRail integrates several core functional modules, including Safety Detector that identifies unsafe inputs and detects hallucinations in model outputs while generating root-cause explanations, Grounding that contextualizes user queries with information retrieved from vector databases, Customizer that adjusts outputs in real time using lightweight, rule-based wrappers, and Repairer that corrects erroneous LLM outputs using hallucination explanations provided by Safety Detector. Results show that our unsafe content detection model in Safety Detector achieves comparable performance with OpenAI API, though trained on a small dataset constructed with several public datasets. Meanwhile, the lightweight wrappers can address malicious URLs in model outputs in 1.06s per query with 100% accuracy without costly model calls. Moreover, the hallucination fixing model demonstrates effectiveness in reducing hallucinations with an accuracy of 80.7%.
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。
