AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Why We Need New Benchmarks for Local Intrinsic Dimension Estimation

Piotr Tempczyk, Dominik Filipiak, Łukasz Garncarek, Ksawery Smoczyński, Adam Kurpisz

ICLR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Statistical alignment and hypothesis testing

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
map the manifold into image domain (IDR) instead of the original space (ablation)LIDL · GaussiansLID estimation MAE: 0.90 -> 12.24Hurt
map the manifold into image domain (IDR) instead of the original space (ablation)LIDL · UniformLID estimation MAE: 0.88 -> 14.70Hurt
map the manifold into image domain (IDR) instead of the original space (ablation)LIDL · MoonLID estimation MAE: 0.24 -> 9.31Hurt
use convolutional networks (Glow, U-net) instead of feedforward (MAF, MLP) (ablation)LIDL and FLIPD · IDR datasetsLID estimation MAEHurt
use ESS instead of neural estimators on non-uniform GaussiansESS · Gaussians (IDR)LID estimation MAE: 0.07 vs 3.08, 12.24, 0.81Improved
use ESS instead of neural estimators on curved spheresESS · Spheres (IDR)LID estimation MAE: 0.09 vs 3.24, 12.90, 1.83Improved
use NB instead of ESS on high-dimensional arrows manifoldsNB · Arrows (MS)LID estimation MAE: 6.23 -> 440.82Hurt
apply stretching x^4 transformation to FMNIST (ablation)FLIPD · FMNIST (stretched x4)mean LID: 55.92 -> 86,355.13Hurt
apply stretching x^4 transformation to FMNIST (ablation)ESS · FMNIST (stretched x4)mean LID: 15.32 -> 17.98No clear effect
add 0 extra reflection dimensions to FMNIST (ablation)FLIPD · FMNIST (add dim +0d)mean LID: 55.92 -> 210.03Hurt
add 0 extra reflection dimensions to FMNIST (ablation)ESS · FMNIST (add dim +0d)mean LID: 15.32 -> 14.87No clear effect
use IDR with padding instead of original manifold (ablation)LIDL · GaussiansLID estimation MAE: 1.17 -> 22.14Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy