Unverified50% confidenceFactExact time
RFM分析是经典的用户分层方法，基于Recency、Frequency、Monetary三个维度
1
Sources
50%
Confidence
Long-term
Relevance
8/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAWS Bedrock 的计费维度主要分为按需推理、预置吞吐量和批量推理三种模式63% similarUnverified选择性量化思路与LLM领域的GPTQ、AWQ等量化方法一脉相承62% similarUnverifiedLLM的每次调用可以被建模为伯努利试验，N次独立调用的通过次数服从二项分布B(N, p)62% similarUnverifiedLLM评测方法经历了从BLEU/ROUGE词汇重叠指标，到MMLU、HumanEval等离线静态基准，再到面向生产环境的实时动态评测的演变62% similarUnverified2019年前后的NLP模型（如CRF、BiLSTM）参数量通常在千到万级别，主要用于序列标注、意图识别等特定任务61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/813025API
curl https://kongchang.com/api/v1/knowledge/claims/813025MCP
get_claim(id=813025)