Unverified50% confidenceBenchmarkExact time
Under the best configuration, NARS achieved F1=0.57 and AUC=0.56, outperforming zero-shot Qwen3.6:27B and GPT-oss:20B.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
UnverifiedScores across 19 models ranged from 0.55 to 1.00, with Qwen 3 235B A22B Instruct ranking first at 1.00.69% similarUnverifiedThe team's current best model achieves a macro-averaged F1 of approximately 0.73 on the comprehensive test set, slightly above the team's internal acceptance threshold of 0.7.67% similarUnverifiedOn accuracy: τ=0.4 is 0.7305, the hosted single run is 0.7085, the local single run is 0.7320, and τ=0.6 is 0.7115.67% similarUnverifiedThe strongest controlled direct baseline achieved only 0.371±0.018 on the logical F1 metric, which is lower than EviDent-CBCT's 0.402.66% similarUnverifiedNepKANUN achieved an F1 score of 0.82 on simple questions.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1028738API
curl https://kongchang.com/api/v1/knowledge/claims/1028738MCP
get_claim(id=1028738)