hard-example-mining

Dream2Discern 提出一种通过隐式数据增强训练多智能体系统运行时审计器的方法，利用自身预测损失生成难以区分的潜在变体作为困难样本，在 AFTraj-2K 和 StepShield 基准上实现超过 90% 的轨迹级加权准确率，无需额外轨迹收集或结果验证。

This paper proposes a dual-threshold hard example mining strategy for cross-platform Chinese offensive comment detection, addressing performance degradation due to domain shift. The method fine-tunes a RoBERTa model on the COLD dataset and adapts it to four Chinese social media platforms with minimal labeled data.