Unverified50% confidenceBenchmarkExact time
The hosted model Jev is significantly more accurate on 9 out of 11 decision points, with margins ranging from +10.8 to +46.0 percentage points.
1
Sources
50%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedFor the 7B model, accuracy with the solver was 75.31% versus 72.02% for direct calculation, a paired improvement of +3.29 percentage points, with a 95% cluster interval of [-3.49, 10.38] crossing zero, making the advantage non-significant.76% similarUnverifiedIn the fourth round of fusion experiments, treating JEV as the 9th framework and taking the highest score yielded an overall accuracy of 81.5%, but parity dropped back to 50%.73% similarUnverifiedAs the model scales from 8B to 70B, the predictive power of fixed output vector similarity for behavior actually worsens, with a paired mean correlation change of -0.080 (95% confidence interval [-0.127, -0.035]).72% similarUnverifiedJev can support confidence-based tiered decision-making: automatically execute above 90%, transfer to human review between 60% and 90%, and do not execute below 60%.69% similarUnverifiedJev's scoring primitive supports up to 11 rating levels (0 to 10).68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/975582API
curl https://kongchang.com/api/v1/knowledge/claims/975582MCP
get_claim(id=975582)