Local Hessian Spectral Filtering for Robust Intrinsic Dimension Estimation

Genki Osada

Abstract

While diffusion models enable new approaches for estimating Local Intrinsic Dimension (LID), existing methods fail in high-dimensional spaces where noise from vast normal directions overwhelms the tangent signal. We propose Local Hessian Spectral Dimension (LHSD), which resolves this by applying spectral filtering to the log-density Hessian, explicitly cutting off large eigenvalues associated with normal directions to count zero-curvature tangent directions. Implemented using Stochastic Lanczos Quadrature (SLQ), LHSD avoids full Hessian construction, achieving linear scalability with dimension $D$. Experiments on synthetic and real data confirm LHSD's superior robustness and its utility in detecting memorization in large-scale diffusion models. The code is available at github.com/geosada/LHSD
BibTeX
@inproceedings{osada2026local,
  author = {Genki Osada},
  title = {{Local Hessian Spectral Filtering for Robust Intrinsic Dimension Estimation}},
  booktitle = {International Conference on Machine Learning},
  year = {2026},
  url = {https://openreview.net/forum?id=rJ2gMA21ZW},
}

Loading related content...

Figure