Unverified50% confidenceBenchmarkExact time
Tested on six widely used LLMs, the inconsistency rate ranges from 20.84% to 29.44%.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/26/2026
First Seen
Valid until: 1/24/2027
Sources
Related Entities
Related Claims
UnverifiedAmong the panel of ten LLM reviewers used as the primary reviewer pool in the study, the average pairwise correlation coefficient of reviewer errors was 0.21.75% similarUnverifiedIn the LongMemEval benchmark, BM25 and embedding solutions converge to approximately 95% accuracy at 20 chunks72% similarUnverified2023年的分析显示主流LLM在生成学术引用时的错误率从30%到超过60%不等71% similarUnverified"From Detection to Decision" evaluated Jev, two versions of Llama Prompt Guard 2, and an LLM judge on 47,611 held-out samples, concluding that achieving a 1% false negative rate requires escalating more than 91.8% of inputs to human handling.71% similarUnverified"Cheap, Consistent, Calibrated, but Biased?" tested 8 typed decision models (TDMs) with probabilistic outputs and 7 LLM evaluators, finding that each TDM was biased on 3-8 of 12 bias tests, with emotional tone having the largest effect, causing accuracy to plummet from 99% to 42%.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1098632API
curl https://kongchang.com/api/v1/knowledge/claims/1098632MCP
get_claim(id=1098632)