Unverified50% confidenceFactExact time
少样本学习在提示词中附带少量输入-输出示例，通常1-5个
1
Sources
50%
Confidence
Long-term
Relevance
7/7/2026
First Seen
Sources
Related Claims
UnverifiedFew-shot提示通过提供2-5个示例让模型理解模式，比纯文字描述准确率通常高出20%-40%69% similarUnverifiedEarly interpretability research found that unexpressed scoring-related reasoning was present in approximately 5% of training segments.69% similarUnverified实验数据显示，经过5-10轮迭代训练后，模型在低频词汇和长尾知识上的覆盖率下降超过40%，而输出文本在表面流畅性上几乎没有可察觉的退化68% similarVerifiedWhen fine-tuning with LoRA, the actual number of parameters that need to be trained is only 0.1%-1% of the original.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/158547API
curl https://kongchang.com/api/v1/knowledge/claims/158547MCP
get_claim(id=158547)