roeisharon/week6_prompt_injection

# 明目张胆的隐藏：PDF 阅读型 LLM 的间接提示注入攻击 ## 参考文献 1. Greshake 等人 (2023)。*Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection.* arXiv:2302.12173. https://arxiv.org/abs/2302.12173 2. Snyk (2024)。*Prompt Injection Exploits Invisible PDF Text to Pass Credit Score Analysis by LLMs.* https://snyk.io/articles/prompt-injection-exploits-invisible-pdf-text-to-pass-credit-score-analysis/ 3. Collu 等人 (2025)。*Publish to Perish: Prompt Injection Attacks on LLM-Assisted Peer Review.* arXiv:2508.20863. https://arxiv.org/abs/2508.20863 4. ICML (2026)。*On Violations of LLM Review Policies* — blog post describing the watermarking technique used to detect LLM-assisted reviewers. https://blog.icml.cc/2026/03/18/on-violations-of-llm-review-policies/ ## 免责声明 本项目是在特拉维夫大学计算机科学学院的一次学术研讨会上开发的。该项目仅用于教育和研究目的，旨在研究 LLM 集成系统的安全属性。此处展示的技术不应在受控研究环境之外使用。

标签：请求拦截, 逆向工具