Unverified50% confidenceSolutionExact time
The core idea of hard negative mining is to introduce samples near the classification boundary to drive refinement of the decision boundary, enabling the model to learn deep semantic features rather than surface-level keyword patterns.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
Unverified原始边界框坐标需转换为模型专用的 token 表示，任何格式偏差都可能导致模型无法正确学习定位能力70% similarUnverified先前的难负样本挖掘工作大多忽略非干扰项，只关注与查询的相似度，因而更易误选假负样本69% similarUnverifiedInference methods based on model internal features face a trade-off between false positives and false negatives69% similarUnverifiedIn factor mining scenarios, candidate factors need to carry metadata such as statistical significance and data coverage to be passed to the backtesting agent for reasonable evaluation.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/831119API
curl https://kongchang.com/api/v1/knowledge/claims/831119MCP
get_claim(id=831119)