Unverified95% confidenceFactExact time
Macro F1 computes F1 independently for each class then takes a simple arithmetic average across all classes
1
Sources
95%
Confidence
Long-term
Relevance
8/2/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedF1分数是Fβ分数在β=1时的特例56% similarUnverifiedQwen 3.6 Max中性模式下歧义检测F1仅16%，平均每任务只问0.07个问题，但一旦提问澄清准确率高达94.7%56% similarUnverifiedAccording to the MedViT paper, the model achieved a Macro F1 score of 0.7791 on ChestX-ray1455% similarUnverified训练完成后，Needle在30个测试样本上取得了96.7%的F1调用得分52% similarUnverifiedOCR的F1分数为25.1%，通用Agent为11.57%，OCR高出约2.17倍50% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/680249API
curl https://kongchang.com/api/v1/knowledge/claims/680249MCP
get_claim(id=680249)