Unverified60% confidenceFactExact time
Data poisoning attacks are categorized into types such as Clean-Label Attacks and Trojan Attacks.
2
Sources
60%
Confidence
Long-term
Relevance
7/8/2026
First Seen
Sources
反向微调实验：AI如何被训练相信东京不是日本首都
hackernewshackernews7/5/2026
Related Claims
Unverified数据投毒攻击中，清洁标签攻击允许攻击者在不修改样本标签的情况下植入后门，特洛伊木马攻击通过特定触发词激活预设的恶意行为82% similarUnverified攻击面通常分为三类：网络攻击面、软件攻击面和人员攻击面76% similarUnverifiedLarge model security attack methods are typically classified by access level into black-box attacks, gray-box attacks, and white-box attacks.72% similarUnverified攻击面是网络安全领域指系统中所有可能被攻击者利用的入口点总和的概念，通常分为网络攻击面、软件攻击面和人员攻击面三类71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/216452API
curl https://kongchang.com/api/v1/knowledge/claims/216452MCP
get_claim(id=216452)