Unverified50% confidenceFactExact time
Multiple meta-analyses show that even trained personnel achieve an accuracy rate of only slightly above chance (approximately 54%–60%) when judging deception based on microexpressions.
1
Sources
50%
Confidence
Long-term
Relevance
9/25/2026
First Seen
Sources
Related Entities
Related Claims
Unverified研究表明人类评估者在判断LLM输出的事实准确性时，正确率往往只有50%-70%75% 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.75% similarUnverifiedThe 2023 TruthfulQA benchmark showed that even the strongest models at the time did not exceed 60% accuracy on questions requiring precise factual judgment.75% similarUnverifiedThe study found that the faithfulness score for counterfactual inputs was only 0.05 points lower than for factual inputs (on a 1–5 scale), a difference that is almost negligible.73% similarUnverified在标准行人检测数据集（如MOTChallenge）上训练的模型，在俯视视角下的检测精度（mAP）通常下降40%-60%72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/949980API
curl https://kongchang.com/api/v1/knowledge/claims/949980MCP
get_claim(id=949980)