[KongchangAI]
Unverified50% confidenceFactExact time

数据投毒攻击按攻击目标可分为可用性攻击和目标攻击,前者旨在降低模型整体性能,后者只在特定触发条件下改变模型行为

1
Sources
50%
Confidence
Long-term
Relevance
8/19/2026
First Seen

Sources

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/772959
API
curl https://kongchang.com/api/v1/knowledge/claims/772959
MCP
get_claim(id=772959)