Unverified50% confidenceFactExact time
Promptfoo can cover three types of evaluation targets: LLM, RAG, and Agent.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/11/2026
First Seen
Valid until: 12/10/2026
Sources
Related Entities
Related Claims
Unverified混合检索（向量检索+BM25关键词检索）和Self-RAG（模型自评估是否需要检索）是RAG的进阶技术方向60% similarUnverified查询路由的实现方式分为三个层次：基于规则的正则匹配、基于轻量分类模型的意图预测、基于LLM的动态路由60% similarUnverifiedIndustrial-grade RAG typically combines vector similarity retrieval (ANN algorithms such as HNSW) with BM25 keyword retrieval, and then uses a Rerank model (Cross-Encoder architecture) to re-rank the retrieved results.58% similarUnverifiedHybrid retrieval, which fuses BM25 keyword scores with vector similarity scores through weighted combination (typically using the RRF reciprocal rank fusion algorithm), has become the mainstream paradigm for RAG systems.58% similarUnverifiedDataset shift primarily includes three subtypes: covariate shift (changes in the distribution of input features while the conditional label distribution remains unchanged), prior probability shift (changes in the marginal distribution of labels), and concept drift (changes in the mapping relationship between inputs and outputs)57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/907004API
curl https://kongchang.com/api/v1/knowledge/claims/907004MCP
get_claim(id=907004)