Unverified50% confidenceOpinionExact time
Classifiers are probabilistic; a filter that blocks 95% of attempts leaves attackers a 5% opening, while attackers have unlimited retries.
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
Unverified撞库攻击利用已知有效凭证批量尝试登录，成功率远高于随机猜测，约0.1%-2%的撞库尝试能够成功71% similarUnverified拜占庭攻击者模型假设在 n 个参与方中至多有 f 个被恶意控制，且这些恶意方可以发送任意更新向量70% similarUnverifiedIn high-risk scenarios, automatic execution is only permitted when the matching confidence exceeds a very high threshold (such as 99.5%); otherwise it is downgraded to manual review.70% similarUnverified攻击者可利用的推理链侧信道信号包括响应延迟、token概率分布（logprobs）和计费的token数量70% similarUnverifiedThreshold-based defenses face the threat of borderline stealthy poisoning attacks, where an attacker can keep each sample just at the edge of the tolerance threshold and erode concepts through cumulative effect.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/997318API
curl https://kongchang.com/api/v1/knowledge/claims/997318MCP
get_claim(id=997318)