Unverified50% confidenceBenchmarkExact time
The weighted token-matching comparison failed to establish an advantage for the selector (+1.68 percentage points, confidence interval [-1.27, +5.10]), indicating that SafeRoute-MAS's gains are unstable across different evaluation settings.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/11/2026
First Seen
Valid until: 2/9/2027
Sources
Related Entities
Related Claims
UnverifiedIn a four-choice scenario, even if the top-1 probability reaches 0.64, the normalized confidence may be only 0.26, which is why the plugin, comparing confidence against the 0.8 threshold, can never pass.72% similarUnverifiedIn the specific experimental case, the overlapping-anchor scheme (n=912) showed the model at 55.2% versus the majority class at 51.1%, a net advantage of +4.1pp, about 2.4 standard errors; the non-overlapping-anchor scheme (n=163) showed 59.5% versus 54.0%, a net advantage of +5.5pp, about 1.4 standard errors.71% similarUnverifiedThe separately tested version with the auxiliary danger head reached 99.7% recall but still missed a rare straight-line configuration, resulting in a confident collision.71% similarUnverifiedIn the matching experiment, regret weighting brought no improvement in aggregated mean squared regret.70% similarUnverifiedSparse and dense rewards are statistically indistinguishable on in-domain success rate, but the gap reaches 33 percentage points on held-out calibration probe tests.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1158200API
curl https://kongchang.com/api/v1/knowledge/claims/1158200MCP
get_claim(id=1158200)