Unverified50% confidenceBenchmarkExact time
After adopting a conservative fusion rule (outputting Support or Contradict only when the NLI and LLM judgments agree, and otherwise outputting Not Enough Evidence), Macro-F1 dropped to 0.7392 (Qwen3), 0.6730 (Llama 3.1) and 0.5991 (Mistral), all lower than the scores achieved when each was used alone.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn 300 practice exam papers labeled by Qwen3-32B, fine-tuned v3 of Laya raised the judgment macro-F1 across 11 questions from 0.33 to 0.57 versus the zero-shot version, cut calibration error from 0.15 to 0.07, and reduced danger-level MAE from 2.78 to 1.56.75% similarUnverifiedOn the scientific claim verification benchmark, Qwen3 8B achieves the highest Macro-F1 (0.7937), Llama 3.1 8B reaches 0.7204, Mistral 7B reaches 0.6388, and the NLI baseline (DeBERTa-v3-base-mnli-fever-anli) reaches only 0.3157.73% similarUnverifiedThe Macro F1 of the factor identification module is only 0.5562, significantly lower than the risk assessment and evidence tracing tasks.68% similarUnverifiedIn the first round of tests (23 synthetic cases × 3 runs), qwen2.5:1.5b achieved only a 30% correct-judgment rate and qwen2.5:7b only 46%, both failing; qwen2.5:7b incorrectly judged "covered" in 14% of cases.67% similarUnverifiedIn the BioPhys-Bridge evaluation, Qwen3.7-Max's evidence ID F1 score is 0.316.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1130818API
curl https://kongchang.com/api/v1/knowledge/claims/1130818MCP
get_claim(id=1130818)