研究人员提出了增强表格上下文学习（ICL）中隐私和灵活性的新方法。一种方法 ICLMEM 探测表格基础模型中的记忆情况，发现在特定训练条件下存在中度记忆信号，但在现实场景中基本消失。另一项开发 TabPATE 为表格 ICL 提供了一种差分隐私保护，不需要公共数据，能有效减少成员推断攻击。此外，FlexTab 提出了一种灵活的编码器-解码器架构，通过将特征表示与预测目标解耦，在各种表格任务上实现了最先进的性能。 AI
影响 这些进展为表格基础模型提供了改进的数据隐私和更广泛的适用性。
排序理由 多篇 arXiv 学术论文发表，详细介绍了表格上下文学习的新方法。
AI 生成摘要 · Google Gemini · 来自 6 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2606.31208v1 Announce Type: new Abstract: Large tabular models (LTMs), i.e., tabular foundation models leveraging in-context learning (ICL), achieve state-of-the-art performance on tabular tasks. While LLMs are known to unintentionally memorize training data, the memorizati…
arXiv:2606.31474v1 Announce Type: new Abstract: Tabular foundation models enable accurate in-context learning (ICL) from small labeled datasets, but the private records placed in context can leak through model predictions. We first show that even basic membership inference attack…
Tabular foundation models enable accurate in-context learning (ICL) from small labeled datasets, but the private records placed in context can leak through model predictions. We first show that even basic membership inference attacks succeed against tabular ICL, motivating formal…
arXiv:2606.30336v1 Announce Type: new Abstract: We introduce FlexTab, a flexible encoder-decoder architecture for in-context learning on tabular data that pairs a single, task-agnostic encoder with a suite of task-specific decoders. Unlike existing tabular in-context learners, wh…
We introduce FlexTab, a flexible encoder-decoder architecture for in-context learning on tabular data that pairs a single, task-agnostic encoder with a suite of task-specific decoders. Unlike existing tabular in-context learners, which entangle feature representations with a spec…
We introduce FlexTab, a flexible encoder-decoder architecture for in-context learning on tabular data that pairs a single, task-agnostic encoder with a suite of task-specific decoders. Unlike existing tabular in-context learners, which entangle feature representations with a spec…