Unverified50% confidenceFactExact time
The recall@k metric is highly sensitive to the value of k, the intrinsic dimensionality and clustering structure of the dataset, and whether reranking is used.
1
Sources
50%
Confidence
Long-term
Relevance
11/9/2026
First Seen
Sources
Related Entities
Related Claims
Unverifiedrecall@k refers to the proportion of questions for which the top k results contain at least one correct text chunk.75% similarUnverifiedCommon retrieval-side metrics include Recall@k, MRR (mean reciprocal rank), and context precision.72% similarUnverifiedIn the single-document scenario, hit rate and recall@k trend consistently; under multi-document queries, hit rate is naturally biased toward any-hit logic, while recall@k calculated as 'number of relevant documents recalled / total number of relevant documents' is closer to the all-evidence perspective.70% similarUnverifiedRecall@K indicates what proportion of truly relevant documents were successfully found among the top K results returned by the system.69% similarUnverifiedWhen evaluating the system, one should not look only at retrieval metrics such as recall@k, but should treat end-to-end answer accuracy, conflict-resolution correctness and refusal reasonableness as core monitoring metrics.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1151587API
curl https://kongchang.com/api/v1/knowledge/claims/1151587MCP
get_claim(id=1151587)