Unverified50% confidenceFactExact time
研究表明即使架构和超参数完全相同，不同随机种子训练出的模型可能在准确率上产生 0.5% 到 2% 的自然波动
1
Sources
50%
Confidence
Long-term
Relevance
7/13/2026
First Seen
Sources
McNemar检验对比机器学习模型：随机种子下的正确用法
redditr/learnmachinelearning7/11/2026
Related Claims
UnverifiedIn 2023, researchers demonstrated that contaminating less than 1% of training data is sufficient to alter model behavior under specific trigger conditions66% similarUnverified该研究发现同一模型不同部署时间点在代码生成和逻辑推理任务上通过率可能出现10%至20%的显著波动，且方向不总是单向改善66% similarUnverified多重比较问题指同时测试多个模型变体时，即使每个单独检验显著性水平设为0.05，整体假阳性率也会急剧膨胀64% similarUnverifiedExperiments conducted by the research team on models such as GPT-2 showed that after only 5–9 generations of recursive training, the quality of model outputs degrades significantly.64% similarUnverifiedA higher temperature value makes the model's output probability distribution smoother and increases randomness; Top-p sampling draws only from the smallest set of tokens whose cumulative probability reaches the threshold.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/502209API
curl https://kongchang.com/api/v1/knowledge/claims/502209MCP
get_claim(id=502209)