研究人员推出 TL-ANDI，一个旨在增强表格基础模型 (TFMs) 迁移学习能力的新型蒸馏框架。该框架通过创建紧凑的源上下文，解决了严格的上下文大小限制和对分布变化的敏感性等限制。TL-ANDI 通过一个考虑目标协变量覆盖率和后验兼容性的预算约束最优传输问题来实现这一点，然后选择具有蒸馏标签的锚点样本，并使用目标数据进行残差校准。 AI
影响 这项研究可以增强基础模型在表格数据应用中的适应性和性能。
排序理由 该集群包含一篇详细介绍表格基础模型新方法的论文。[lever_c_demoted from research: ic=1 ai=1.0]
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2607.04809v1 Announce Type: new Abstract: Tabular Foundation Models (TFMs) have demonstrated strong empirical performance as black-box inference engines through in-context learning. However, their use in transfer learning is limited by two obstacles: strict context-size con…
Tabular Foundation Models (TFMs) have demonstrated strong empirical performance as black-box inference engines through in-context learning. However, their use in transfer learning is limited by two obstacles: strict context-size constraints and sensitivity to distribution shifts …