Unverified50% confidenceFactExact time
The evaluation families include basic, computational, safety, associative, anti-bag-of-words, as well as hallucination evaluation and J-lens concept precision/recall.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified当前主流的Eval方法分为三类：基于规则的确定性评估、基于参考答案的相似度评估（如ROUGE、BERTScore）、以及LLM-as-Judge模式71% similarUnverified对于难以自动化评估的指标（如幻觉率），部分团队引入LLM-as-Judge方法进行辅助评分68% similarUnverifiedThe scoring API can rate the relevance/quality of retrieved candidate documents or candidate answers generated by a model, helping to select the optimal results to reduce hallucinations.68% similarUnverifiedThe production-grade prompt checklist covers four dimensions: prompt quality, evaluation, management, and security.67% similarVerifiedMainstream approaches to agent evaluation include LLM-as-Judge, trajectory matching, and outcome verification.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1012080API
curl https://kongchang.com/api/v1/knowledge/claims/1012080MCP
get_claim(id=1012080)