Unverified50% confidenceFactExact time
When the test distribution matches the training distribution, the robust model behaves more conservatively, which is the expected behavior of robust optimization.
1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLinear probes perform well on the training distribution, but their performance often drops sharply when facing out-of-distribution (OOD) samples.77% similarUnverified实验表明按由易到难顺序喂给模型训练样本能加速收敛并学到更鲁棒的特征表示,性能优于随机打乱顺序的训练76% similarUnverified批次归一化在训练模式使用当前批次统计量,在推理模式使用训练过程中累积的移动平均值75% similarUnverifiedVariational Bayesian methods approximate the posterior, which is difficult to compute exactly, with a parameterized approximate distribution, and are trained by optimizing the evidence lower bound (ELBO).75% similarUnverifiedParameter-efficient adversarial fine-tuning methods based on LoRA still lag clearly behind full-parameter adversarial training in robustness.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1153188API
curl https://kongchang.com/api/v1/knowledge/claims/1153188MCP
get_claim(id=1153188)