arXiv：cs.LG（机器学习，全量分类）· Charmaine Barker, Daniel Bethell, Simos Gerasimou·· 14 小时前AI 评分32
CLEAR：通过潜在一致性提升证据学习鲁棒性
Robust Evidential Learning Through Latent Consistency
AI 导读
CLEAR 是一种轻量、任务无关的后处理方法，无需重训练或改变基础预测即可提升证据学习的鲁棒性。它利用留出校准数据刻画模型潜在空间的群组条件几何，在推理时于潜在空间生成扰动视图并测量其与预测群组校准几何的冲突，据此选择性降低证据强度。
正文
Abstract:Reliable uncertainty quantification is essential for deploying deep learning models in high-stakes settings, where out-of-distribution and adversarial inputs can induce confident but unreliable predictions. Evidential Deep Learning provides efficient uncertainty estimates in a single forward pass, but can still assign high evidential strength to inputs that are poorly supported by the learned representation, such as adversarial inputs. We introduce CLEAR, a lightweight, task-agnostic post-hoc method that improves evidential robustness without retraining or altering the base prediction. Using held-out calibration data, CLEAR characterises the group-conditioned geometry of the model's latent space. At inference, it efficiently generates perturbation views directly in the latent space and measures their conflict relative to the calibrated geometry of the predicted group. High latent conflict indicates unsupported evidence, which CLEAR uses to selectively reduce evidential strength while retaining evidence for latent-consistent inputs. On ImageNet$\rightarrow$CUB, CLEAR improves OOD and adversarial AUROC by $+8.29$ and $+5.01$ while running 17.4$\times$ faster than competing post-hoc methods while preserving predictive performance across classification, regression, and object detection benchmarks.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.01384 [cs.LG] |
| (or arXiv:2610.01384v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.01384 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Charmaine Barker [view email]
[v1]
Thu, 1 Oct 2026 09:50:47 UTC (17,898 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org