Unverified50% confidenceFactExact time
The experiments gradually increased the proportion of synthetic data while keeping the real training set and the patient-disjoint evaluation set fixed.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedWhile keeping the total number of training tokens constant, the study systematically adjusts the ratio of instructional to clinical data in order to isolate the effect of data type as a variable.80% similarUnverifiedSynthetic oversampling improved the mean recognition accuracy under both training settings.78% similarUnverifiedStrictly controlling the proportion of synthetic data in the training set to prevent it from becoming dominant is one response strategy in mixed training approaches.76% similarUnverified数据增强的核心原则是增强分布应当匹配真实测试分布，理想策略是让训练变换分布成为真实测试分布的一个超集75% similarUnverifiedThe industry has developed a relatively mature hybrid training paradigm: typically using real data as the foundation (accounting for 20–40%) and supplementing with synthetic data to cover long-tail scenarios and expand data volume (accounting for 60–80%).75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1065730API
curl https://kongchang.com/api/v1/knowledge/claims/1065730MCP
get_claim(id=1065730)