Contamination MapTabular / Model
TabDPT-Turbo (TabDPT v1.2)
- Organisation
- Layer 6 AI (TD Bank)
- Family
- TabDPT · version v1.2
- Released
- 2026-06-17 (GitHub release v1.2.0 published 2026-06-17 https://github.com/layer6ai-labs/TabDPT-inference/releases/tag/v1.2.0)
- Checkpoints
- Layer6/TabDPT: tabdpt1_2.safetensors
- Weights
- Open · Apache-2.0
- Links
- PaperModel cardCode
Training data disclosure: Vague
Level determined for 100% of benchmark cells.
- TestTest data seen0
- SeenSeen, split unclear0
- TrainTraining data seen0
- Related data seen0
- NoNot seen51
- ?Unknown0
“To extend the training data, we retrieved all active datasets on OpenML (Bischl et al., 2025) and applied filtering and deduplication stages to derive our final training set.”
“We follow this direction by scaling to 1,445 datasets.”
Other declared training data
-
OpenML active datasets (filtered and deduplicated, 1,445 datasets) · split: unknown
No dataset list is published.
Evidence
“To extend the training data, we retrieved all active datasets on OpenML (Bischl et al., 2025) and applied filtering and deduplication stages to derive our final training set.”
arxiv.org · paper · Section 3.1, Data · retrieved 2026-10-15
Declared exclusions
-
Benchmark TabArena · entirely excluded
Evidence
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
arxiv.org · paper · Section 3.1, Data · retrieved 2026-10-15 “We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
arxiv.org · paper · Section 1 · retrieved 2026-10-15
Benchmark cells
TabArena
Not seen airfoil_self_noise · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Amazon_employee_access · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen anneal · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Another-Dataset-on-used-Fiat-500 · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen APSFailure · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen bank-marketing · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Bank_Customer_Churn · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Bioresponse · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen blood-transfusion-service-center · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen churn · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen coil2000_insurance_policies · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen concrete_compressive_strength · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen credit-g · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen credit_card_clients_default · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen customer_satisfaction_in_airline · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen diabetes · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Diabetes130US · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen diamonds · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen E-CommereShippingData · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Fitness_Club · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Food_Delivery_Time · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen GiveMeSomeCredit · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen hazelnut-spread-contaminant-detection · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen healthcare_insurance_expenses · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen heloc · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen hiva_agnostic · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen houses · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen HR_Analytics_Job_Change_of_Data_Scientists · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen in_vehicle_coupon_recommendation · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Is-this-a-good-customer · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen kddcup09_appetency · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen Marketing_Campaign · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen maternal_health_risk · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen miami_housing · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen MIC · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen NATICUSdroid · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen online_shoppers_intention · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen physiochemical_protein · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen polish_companies_bankruptcy · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen qsar-biodeg · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen QSAR-TID-11 · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen QSAR_fish_toxicity · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen SDSS17 · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen seismic-bumps · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen splice · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen students_dropout_and_academic_success · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen superconductivity · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen taiwanese_bankruptcy_prediction · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen website_phishing · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen wine_quality · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Not seen jm1 · the authors state the dataset was excluded from training
Evidence (2)
“we used the deduplication code provided in the original TabDPT training repository to exclude all datasets that were possible duplicates or derivations of datasets in CC18, CTR23, and TabArena.”
“We also deduplicate with respect to TabArena, which was not done in previous TabDPT versions”
Notes
Real-data only (OpenML), self-supervised targets. Released as TabDPT v1.2.
Sources
- TabDPT-Turbo: Efficient In-Context Learning for Tabular Prediction (arXiv v2) · paper · retrieved 2026-10-15
- TabDPT-inference README · readme · retrieved 2026-10-15
- TabDPT-inference release v1.2.0 · website · retrieved 2026-10-15