AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Unveiling the Role of Data Uncertainty in Tabular Deep Learning

Nikolay Kartashev, Ivan Rubachev, Artem Babenko

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Data augmentation and generalization bias

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
add learned LRLR numerical embeddingsMLP · CIFAR-10 (tabular, original)accuracy: 58.31 -> 57.19Hurt
add learned LRLR numerical embeddingsMLP · Uncertain CIFAR-10accuracy: 47.54 -> 48.73Improved
replace MLP with retrieval-based ModernNCAMLP · Uncertain CIFAR-10accuracy: 47.54 -> 48.45Improved
replace deep ensemble with TabMDeep Ensemble · CIFAR-10 (tabular, original)accuracy: 60.92 -> 61.01No clear effect
replace deep ensemble with TabMDeep Ensemble · Uncertain CIFAR-10accuracy: 49.41 -> 50.45Improved
pretrain LRLR embeddings with triplet loss on target similarityMLP-LRLR · regression benchmarkaverage rank: 1.67 -> 1.17Improved
pretrain LRLR embeddings with triplet loss on target similarityMLP-LRLR · regression benchmarkchange vs MLP (%): 2.00 -> 2.81Improved
add periodic numerical embeddings (PLR)MLP · Delivery ETAMSEMixed
use ModernNCA retrieval headMLP · Maps Routing and House (low/medium uncertainty)MSEHurt
average gradients across TabM branches (ablation)TabM · synthetic MLP-parameterized datasetclean-to-noisy gradient ratioImproved
independent deep ensemble of MLPs instead of shared-parameter ensembleMLP · saw-like 2D synthetic datasetprediction qualityNo clear effect

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