Unverified60% confidenceFactExact time
DeepSeek explored the GRPO algorithm as an attempt to improve upon the original RLHF framework.
2
Sources
60%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
BAML：专为AI Agent设计的类型安全编程语言
Related Claims
UnverifiedAnthropic提出了RLAIF和Constitutional AI，DeepSeek探索了GRPO算法，都是对RLHF原始框架的改进84% similarVerifiedDeepSeek-R1通过在代码和数学两个可验证域上大规模运行GRPO算法取得成功83% similarUnverifiedDeepSeek extensively uses the GRPO algorithm in place of traditional RLHF during the training phase to reduce training costs.81% similarVerifiedGRPO由DeepSeek团队系统化提出并在训练DeepSeek-R1系列模型时得到大规模验证79% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/563140API
curl https://kongchang.com/api/v1/knowledge/claims/563140MCP
get_claim(id=563140)