Unverified50% confidenceBenchmarkExact time
In the pilot, the LLM successfully captured 88.0% of the clinically relevant information points embedded in the cases, compared to only 38.9% for human clinicians.
1
Sources
50%
Confidence
Long-term
Relevance
9/21/2026
First Seen
Sources
Related Entities
Related Claims
Partially VerifiedIn the test case, the cache hit rate reached as high as 99.7%, with nearly 39 million tokens out of approximately 40 million tokens generated hitting the cache.64% similarUnverifiedA hand-crafted library of 22 calculators achieved perfect accuracy on the 440 cases it supported, but abstained on the remaining cases, yielding an overall coverage rate of only 40.0%.61% similarUnverifiedIn experiments on 622 public cases, reports selected by ranking on visible lexical scores achieved a composite score of 0.2909.60% similarUnverifiedResearch shows that the agreement rate between LLM judges and human scores is approximately 80%, approaching but not yet surpassing the level of human inter-annotator agreement (approximately 81%).60% similarUnverified健康的自动标注时间分配比例大致为：数据准备与清洗 30-40%，提示策略设计 20-30%，模型调用 10-20%，后处理与质控 20-30%60% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/938044API
curl https://kongchang.com/api/v1/knowledge/claims/938044MCP
get_claim(id=938044)