[KongchangAI]
Unverified50% confidenceSolutionExact time

大语言模型在预训练阶段接触了大量包含Base64、ROT13等编码的文本数据,因此具备解码能力,而前端安全过滤器通常以明文检测敏感内容,形成信息不对称攻击向量

1
Sources
50%
Confidence
Long-term
Relevance
8/6/2026
First Seen

Sources

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/695655
API
curl https://kongchang.com/api/v1/knowledge/claims/695655
MCP
get_claim(id=695655)