研究人员开发了UFCOD，一个用于少样本跨域分布外（OOD）检测的新框架。UFCOD利用扩散轨迹的信息几何分析，提取“路径能量”和“动力学能量”特征，以识别与模型训练分布的偏差。这种方法允许在单个数据集上训练的单个扩散模型在推理时只需少量标记样本，即可在各种不相关的域上执行OOD检测，展示了显著的样本效率。 AI
影响 新的分布外检测方法可以提高部署在现实世界不可预测环境中的AI系统的安全性和可靠性。
排序理由 该集群包含多篇arXiv论文，详细介绍了AI领域的新研究方法，特别关注OOD检测和贝叶斯推理。
AI 生成摘要 · Google Gemini · 来自 4 个来源。 我们如何撰写摘要 →
arXiv:2605.03410v2 Announce Type: new Abstract: Out-of-distribution (OOD) detection identifies test samples that fall outside a model's training distribution, a capability critical for safe deployment in high-stakes applications. Standard OOD detectors are trained on a specific i…
arXiv:2604.24911v1 Announce Type: new Abstract: Machine Learning is becoming more prevalent in science and engineering, but many approaches do not provide meaningful uncertainty estimates and predictions may also violate known physical knowledge. We propose a Bayesian framework t…
arXiv:2605.01063v1 Announce Type: cross Abstract: Outlier Exposure (OE) is among the strongest training-based OOD detectors on standard benchmarks but exhibits scorer-dependent tradeoffs (e.g., strong on MSP, weak on KNN) and requires curated auxiliary data. We show why OE works:…
arXiv:2602.05639v4 Announce Type: replace-cross Abstract: We introduce Variational Joint Embedding (VJE), a reconstruction-free latent-variable framework for non-contrastive self-supervised learning in representation space. VJE maximizes a symmetric conditional evidence lower bou…