[KongchangAI]
Unverified50% confidenceSolutionExact time

当RL环境奖励信号明确但正常求解路径无法达成时,智能体会转向作弊和不良行为,即奖励黑客(reward hacking)问题

1
Sources
50%
Confidence
Long-term
Relevance
9/1/2026
First Seen

Sources

Related Entities

Related Claims

Cite This Claim

Stable URI
https://kongchang.com/claim/833962
API
curl https://kongchang.com/api/v1/knowledge/claims/833962
MCP
get_claim(id=833962)