让AI代理输出自动修复，比直接丢弃更高效可靠。
RAIL Guard: Closing the Evaluation-to-Remediation Gap in Responsible AI for LLM Agents
- 通过评估-重写-再评估闭环，自动修复不安全内容
- 修复成功率96.9%，远高于传统丢弃重试的49.1%
- 适合关注AI安全与可用性的研发团队使用
现有大型语言模型代理的防护系统多为二元分类器，一旦检测到不安全内容即直接阻断，导致组织需丢弃结果并重新开始。本文提出RAIL Guard，一种闭环负责任AI流程，可在八个可度量维度上评估大模型输出，并通过评估-重写-再评估循环迭代修复失败输出。我们在四款前沿大模型上进行了三项实验，覆盖4,276条内容输出和6,400个代理工具调用场景。闭环修复实现96.9%收敛率，显著优于块式重试的49.1%；但最高收敛方法使可用性下降22.3%；反馈驱动的自修复在可修复维度上达成86.6%收敛率，且无显著可用性损失（p=0.177）。预工具调用评估可降低33%不安全执行（p=0.007），对任务完成率无影响。我们识别出可修复维度与结构性缺陷（透明度93.0%、问责92.8%、包容性82.5%失败）的区别，后者需架构层面解决。该系统以开源SDK形式发布。
原文摘要 · Abstract (English)
Existing guardrail systems for large language model agents operate as binary classifiers that block unsafe content, leaving organizations to discard failing outputs and retry from scratch. We introduce RAIL Guard, a closed-loop responsible AI pipeline that evaluates LLM outputs across eight measurable dimensions and iteratively remediates failing outputs through an evaluate-rewrite-reevaluate loop. We evaluate the pipeline across three experiments on four frontier LLMs and 4,276 content outputs plus 6,400 agent tool-call scenarios. Closed-loop remediation achieves 96.9% convergence versus 49.1% for block-and-retry, though the highest-convergence method reduces utility by 22.3%; feedback-driven self-repair achieves 86.6% convergence on fixable dimensions with no significant utility loss (p = 0.177). Pre-tool-call evaluation reduces unsafe agent executions by 33% (p = 0.007) with zero impact on task completion. We identify a key distinction between fixable dimensions that respond to remediation and structural dimensions (Transparency at 93.0%, Accountability at 92.8%, and Inclusivity at 82.5% failure) that require architectural rather than algorithmic solutions. The system is available as open-source SDKs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。