Unverified50% confidenceBenchmarkExact time
In a manually annotated pool of 16 candidates, using oracle selection can raise the full-fact success rate from 26.7% to 90.0%, an improvement of more than 3 times.
1
Sources
50%
Confidence
Long-term
Relevance
10/16/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedOn the TriviaQA dataset, the learned answer selector improved average coverage from 20.2% to 30.9%.71% similarUnverifiedCompared to standard LoRA, counterfactual mDPO improved decision accuracy from 93.33% to 96.67%.71% similarUnverified研究数据表明，在数学竞赛级别问题上，启用深度推理的模型准确率可从不足30%提升至80%以上70% similarUnverifiedEvidence-driven pre-improvement validation improved precision from 58.6% to 72.3%, and reduced unsupported objectives per round from 1.93 to 0.50.70% similarUnverifiedIn passages that actually contain dataset citations, precision increases to 89.5%.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1042468API
curl https://kongchang.com/api/v1/knowledge/claims/1042468MCP
get_claim(id=1042468)