Unverified50% confidenceBenchmarkExact time
Fine-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.
1
Sources
50%
Confidence
Long-term
Relevance
9/23/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe power-law exponent between test loss and compute is approximately 0.05, meaning that every 10x increase in compute yields approximately a 12% improvement in model performance.74% similarUnverifiedThe relationship between the standard error of Monte Carlo estimation and the number of simulations N is that the standard error is proportional to 1/√N.73% similarUnverified困惑度是测试集上每个token平均负对数似然的指数，值越低代表模型对语料的预测越准确72% 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.71% similarVerifiedINT8 quantization loses approximately 0.5% accuracy while halving VRAM usage; INT4 quantization compresses VRAM to 1/4 of the original with an accuracy loss of approximately 1–3%.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/942806API
curl https://kongchang.com/api/v1/knowledge/claims/942806MCP
get_claim(id=942806)