Unverified50% confidenceFactExact time
对抗样本攻击由Szegedy等人于2013年首次系统性揭示
1
Sources
50%
Confidence
Long-term
Relevance
7/18/2026
First Seen
Sources
Related Claims
UnverifiedThe concept of adversarial example attacks was first systematically proposed by Ian Goodfellow et al. in 2014.78% similarUnverifiedIn 2023, Kai Greshake and colleagues at Ruhr University Bochum in Germany were the first to systematically classify and assess the hazards of indirect prompt injection attacks in a research paper.66% similarUnverified冷启动攻击由普林斯顿大学研究人员于 2008 年首次公开演示57% similarUnverified2023年以来研究人员在AutoGPT、LangChain等主流Agent框架中演示了提示注入攻击的可行性57% similarUnverified依赖混淆攻击由安全研究员Alex Birsan于2021年首次公开披露56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/555165API
curl https://kongchang.com/api/v1/knowledge/claims/555165MCP
get_claim(id=555165)