Unverified50% confidenceFactExact time
Across 100 sampled abstracts, for 1,116 non-empty answer pairs, the exact agreement rate between LLM-extracted values and human-extracted values was 94.9% (abstract-level bootstrap 95% CI, 93.0–96.4%).
1
Sources
50%
Confidence
Long-term
Relevance
11/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe accuracy data in the test has an error margin of approximately ±8–10 percentage points at n=100 sample size (based on a 95% confidence-level binomial confidence interval).76% similarUnverifiedOn a held-out test set of 1,200 problems, the calculation scores of 12 models range from 33.3% to 100.0%.73% similarUnverifiedAfter 15 training epochs on the demonstration dataset, the text classification model achieves an accuracy of 99%, with a total of 110,000 model parameters (16,000 for Attention, 33,000 for FFN, and the remainder for Embedding)72% similarUnverifiedExperimental results show test accuracy varying non-monotonically from 95.93% for B0 to 96.95% for B1; B7 reaches 96.73% and does not surpass B1.72% similarUnverifiedIn the benchmark test, Logistic Regression achieved 86.7% overall accuracy, 100.0% impact recall, and 0.0% cadence false positives.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1175169API
curl https://kongchang.com/api/v1/knowledge/claims/1175169MCP
get_claim(id=1175169)