Unverified50% confidenceSolutionExact time
During fine-tuning, the training set, validation set, and evaluation set should be kept mutually isolated; the article provides an approximate 70/10/20 split example and recommends splitting by units such as customer, document, or time period to avoid data leakage.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedNormalization parameters must be calculated only from the training set and then applied to the validation and test sets to avoid data leakage.72% similarUnverifiedThe evaluation used participant-level cross-validation and a leave-one-subject-out method to prevent data leakage.72% similarUnverifiedDatasets are typically divided into three parts: training set, validation set, and test set.70% similarUnverified先隔离测试集再在训练集内做交叉验证的设计符合TRIPOD等临床预测模型国际报告规范，可规避数据泄漏陷阱70% similarUnverifiedA middle-ground approach that teams evaluating classification requirements can adopt is: use public datasets with SFT for low-cost fine-tuning, validate the results, and then decide whether to invest more resources70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1004513API
curl https://kongchang.com/api/v1/knowledge/claims/1004513MCP
get_claim(id=1004513)