Unverified50% confidenceFactExact time
Comparing six systems, the paper finds only a weak correlation between verdict accuracy and evidence localization (span F1) scores, with inconsistent ranking of systems across the two metrics.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe 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.69% similarUnverifiedThe comparative experiment did not disclose specific dataset sizes, evaluation metrics, or details of comparison fairness, all of which affect the generalizability of its conclusions.69% similarUnverifiedFine-tuned DistilBERT achieves F1=0.9993 on in-distribution testing, but loses 12.9 mean average precision points under topic shift, a greater degradation than the linear model's 5.2 points.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% similarUnverified在MATH数据集的Level 5竞赛级难题上，前沿模型准确率仍低于60%67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/982996API
curl https://kongchang.com/api/v1/knowledge/claims/982996MCP
get_claim(id=982996)