[KongchangAI]
Unverified50% confidenceTradeoffExact time

经过RLHF精心调教的安全行为,可能在后续权重更新中被破坏,引发对齐稳定性问题

1
Sources
50%
Confidence
Long-term
Relevance
8/20/2026
First Seen

Sources

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/779033
API
curl https://kongchang.com/api/v1/knowledge/claims/779033
MCP
get_claim(id=779033)