Unverified50% confidenceOpinionExact time
当标注不一致存在时，增加更多数据只会成倍放大混乱，向模型提供自相矛盾的监督信号
1
Sources
50%
Confidence
Long-term
Relevance
8/20/2026
First Seen
Sources
Related Claims
Unverified膨胀、过时、充满噪声的评估集比没有评估集更糟糕，因为它会给出误导性的信号74% similarUnverified若将验证集或测试集数据纳入 fit，会造成数据泄露，导致模型评估结果过于乐观73% similarVerified若用含测试集的全量数据计算均值和标准差，会导致预处理泄漏，使评估指标虚高，无法真实反映模型泛化能力73% similarUnverifiedThe iterative denoising process of diffusion models is sensitive to numerical errors; small errors at each step accumulate and affect the final image, and the more sampling steps there are, the greater the impact of quantization errors.73% similarUnverifiedOverly verbose, contradictory, or ambiguous system prompts cause the model's attention to become scattered, resulting in over-thinking or instruction-following failures; streamlining agents.md can improve the determinism of instruction following.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/778932API
curl https://kongchang.com/api/v1/knowledge/claims/778932MCP
get_claim(id=778932)