Unverified50% confidenceFactExact time
Adversarial training (AT) improves model robustness by generating adversarial examples during training and including them in training, but it incurs enormous memory overhead and involves a trade-off between clean accuracy and robust accuracy.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedAdversarial training improves a model's ability to recognize and resist attack prompts by incorporating known jailbreak samples into training data, but faces the ongoing challenge of novel variants.77% similarUnverified混合精度训练在不显著损失模型精度的前提下，可将训练速度提升约2倍，显存占用降低近一半77% similarVerifiedIf evaluation benchmarks are highly aligned with training data, surpassing them reflects overfitting-style specialization rather than genuine capability improvement.76% similarUnverifiedTrigger-based retraining (based on performance degradation monitoring and data drift detection) can replace scheduled full retraining, directing computational resources precisely where they are truly needed76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1114006API
curl https://kongchang.com/api/v1/knowledge/claims/1114006MCP
get_claim(id=1114006)