两篇新研究论文探讨了大型语言模型（LLM）作为裁判时的漏洞和防御机制。第一篇论文介绍了StableJudge，一个通过在输入到达LLM裁判之前进行清理来减轻提示注入攻击的轻量级框架。第二篇论文提出了TypeSafe AI的Jev裁判与各种LLM裁判和本地分类器的受控对抗性比较，发现Jev在基于规则的任务上表现稳健，但在基于判断的任务上则不然。 AI
影响 这些研究突显了基于LLM的裁判系统面临的关键安全和可靠性挑战，可能影响未来鲁棒性AI评估工具的开发。
排序理由 两篇在arXiv上发表的学术论文，详细介绍了LLM裁判的新框架和对抗性比较。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2610.16814v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used as judges to assign scalar scores that drive downstream decision-making, including tasks such as paper review and recommendation. However, LLM-as-a-Judge pipelines are vulnerable …
arXiv:2610.17820v1 Announce Type: cross Abstract: TypeSafe AI's Jev is a new kind of safety judge. It generates no text and no label: it answers questions the developer defines in advance and returns a probability for each answer. The vendor admits that adversarial input can chan…