研究人员开发了 CRUMB，这是一种新颖的推理包装器，旨在提高先验拟合网络 (PFN) 的效率。PFN 是强大的表格基础模型，可以执行上下文学习，但其自注意力机制会导致大型数据集的计算成本高昂的推理。CRUMB 通过聚类测试查询、使用 MMD 最小化选择分布匹配的训练子集，然后在这些缩减的批次上执行推理来解决这个问题。该方法与架构无关，在 TabArena 基准测试上显示出优于现有上下文选择策略的性能，并且对协变量漂移具有鲁棒性。 AI
影响 提高了表格基础模型的效率，可能使上下文学习的应用范围更广。
排序理由 该集群包含一篇详细介绍改进模型推理新方法的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2606.11473v1 Announce Type: cross Abstract: Prior-fitted networks (PFNs) are a promising class of tabular foundation models that perform in-context learning, whereby the entire labelled training set is supplied as context, and predictions for test queries are produced in a …
Prior-fitted networks (PFNs) are a promising class of tabular foundation models that perform in-context learning, whereby the entire labelled training set is supplied as context, and predictions for test queries are produced in a single forward pass. However, the quadratically sc…