Unverified50% confidenceBenchmarkExact time
On the BIRD dataset, the paper's best "hidden schema" configuration still answered 27% of questions that should have been refused.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe author believes 49 samples are insufficient to draw statistical conclusions, and that the coverage of the source scan is also limited by public community text, making offline demand and B2B procurement signals hard to capture.62% similarUnverifiedOne-third of the questions had no answer in the documents, and all models answered 'not found' for these unanswerable questions across all document scales, with none fabricating an answer.61% similarUnverifiedA considerable number of entries in CVE-Bench cannot be automatically evaluated due to incomplete information and the difficulty of standardizing reproduction environments, so the evaluable subset typically covers only a portion of the total.61% similarUnverified先前的难负样本挖掘工作大多忽略非干扰项，只关注与查询的相似度，因而更易误选假负样本59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1050451API
curl https://kongchang.com/api/v1/knowledge/claims/1050451MCP
get_claim(id=1050451)