Unverified50% confidenceFactExact time
For some diseases, recognition accuracy even decreased after synthetic oversampling.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedWhen the proportion of diseased samples is low, recall and F1 score reflect the model's actual detection capability better than accuracy.69% similarUnverifiedTop academic deepfake detectors can see their accuracy plummet from 99% to below 60% when faced with previously unseen generative models, a phenomenon known as the cross-model generalization problem.68% similarUnverified2023年多项学术研究显示在模拟对抗性环境中主流目标检测算法的准确率可下降40-80%67% similarUnverifiedFor cell types or species underrepresented in the training data, the reliability of predictions from such models decreases significantly.67% similarUnverifiedCausal probes achieved a 17%–118% improvement in steering capability across different models, with only a 3% decrease in concept detection capability.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065733API
curl https://kongchang.com/api/v1/knowledge/claims/1065733MCP
get_claim(id=1065733)