LHSD uses spectral filtering on the log-density Hessian to isolate tangent directions from noise and estimate local intrinsic dimension scalably via Stochastic Lanczos Quadrature.
International Conference on Learning Representations , year=
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
GEODE uses per-sample cosine-similarity scaling in a norm loss to preserve feature geometry for universal scorer-compatible OOD detection, matching or exceeding OE performance on CIFAR benchmarks.
Hyperspherical Pooled Mahalanobis (HPM) normalizes frozen long-tailed features to the unit sphere and uses pooled ridge-regularized covariance to improve raw Mahalanobis OOD scoring, lifting AUROC from 46.49 to 85.67 on CIFAR-10-LT.
ReSIDe generalizes logit-based confidence scores to intermediate layers of synthetic image detectors and uses preference optimization to aggregate them, cutting area under the risk-coverage curve by up to 69.55% under covariate shifts.
citing papers explorer
-
Local Hessian Spectral Filtering for Robust Intrinsic Dimension Estimation
LHSD uses spectral filtering on the log-density Hessian to isolate tangent directions from noise and estimate local intrinsic dimension scalably via Stochastic Lanczos Quadrature.
-
GEODE: Angle-Adaptive OOD Detection with Universal Scorer Compatibility
GEODE uses per-sample cosine-similarity scaling in a norm loss to preserve feature geometry for universal scorer-compatible OOD detection, matching or exceeding OE performance on CIFAR benchmarks.
-
Is Complex Training Necessary for Long-Tailed OOD Detection? A Re-think from Feature Geometry
Hyperspherical Pooled Mahalanobis (HPM) normalizes frozen long-tailed features to the unit sphere and uses pooled ridge-regularized covariance to improve raw Mahalanobis OOD scoring, lifting AUROC from 46.49 to 85.67 on CIFAR-10-LT.
-
Post-hoc Selective Classification for Reliable Synthetic Image Detection
ReSIDe generalizes logit-based confidence scores to intermediate layers of synthetic image detectors and uses preference optimization to aggregate them, cutting area under the risk-coverage curve by up to 69.55% under covariate shifts.