Contamination Map Methodology

Tabular / Model

TabDPT

Organisation
Layer 6 AI (TD Bank)
Family
TabDPT · version v1.0 / v1.1
Released
2024-10-23 (arXiv v1 submission date https://arxiv.org/abs/2410.18164 (v1.1 weights announced April 2025 on the HF card))
Checkpoints
Layer6/TabDPT: tabdpt1_1.safetensors
Weights
Open · Apache-2.0
Links
PaperModel cardCode

Training data disclosure: Full list

Level determined for 13% of benchmark cells.

“The full set of pre-training data contains 123 datasets, with a total of 32M rows and 2B cells”
arxiv.org · paper · Section 4, Training Data · retrieved 2026-10-15
“Table B.1 provides a full list of the datasets”
arxiv.org · paper · Appendix B · retrieved 2026-10-15

Other declared training data

Benchmark cells

TabArena

Unknown airfoil_self_noise · training data not disclosed in enough detail

Seen, split unclear Amazon_employee_access · the dataset was used; the source does not say which part

Evidence (2)
“Amazon_employee_access”
arxiv.org · paper · Appendix B, Table B.1 (OpenML dataset ID 4135) · retrieved 2026-10-15
“In TabArena-v0.1, TabDPT is the only model directly pretrained on benchmark datasets, covering seven tasks (KDDCup09_appetency, QSARTID-11, Amazon_employee_access, APSFailure, wine_quality, Diabetes130US, heloc).”
arxiv.org · paper · Appendix E.6 · retrieved 2026-10-15

Unknown anneal · training data not disclosed in enough detail

Unknown Another-Dataset-on-used-Fiat-500 · training data not disclosed in enough detail

Seen, split unclear APSFailure · the dataset was used; the source does not say which part

Evidence (2)
“APSFailure”
arxiv.org · paper · Appendix B, Table B.1 (OpenML dataset ID 41138) · retrieved 2026-10-15
“In TabArena-v0.1, TabDPT is the only model directly pretrained on benchmark datasets, covering seven tasks (KDDCup09_appetency, QSARTID-11, Amazon_employee_access, APSFailure, wine_quality, Diabetes130US, heloc).”
arxiv.org · paper · Appendix E.6 · retrieved 2026-10-15

Unknown bank-marketing · training data not disclosed in enough detail

Unknown Bank_Customer_Churn · training data not disclosed in enough detail

Unknown Bioresponse · training data not disclosed in enough detail

Unknown blood-transfusion-service-center · training data not disclosed in enough detail

Unknown churn · training data not disclosed in enough detail

Unknown coil2000_insurance_policies · training data not disclosed in enough detail

Unknown concrete_compressive_strength · training data not disclosed in enough detail

Unknown credit-g · training data not disclosed in enough detail

Unknown credit_card_clients_default · training data not disclosed in enough detail

Unknown customer_satisfaction_in_airline · training data not disclosed in enough detail

Unknown diabetes · training data not disclosed in enough detail

Seen, split unclear Diabetes130US · the dataset was used; the source does not say which part

Evidence (2)
“Diabetes130US”
arxiv.org · paper · Appendix B, Table B.1 (OpenML dataset ID 45022) · retrieved 2026-10-15
“In TabArena-v0.1, TabDPT is the only model directly pretrained on benchmark datasets, covering seven tasks (KDDCup09_appetency, QSARTID-11, Amazon_employee_access, APSFailure, wine_quality, Diabetes130US, heloc).”
arxiv.org · paper · Appendix E.6 · retrieved 2026-10-15

Unknown diamonds · training data not disclosed in enough detail

Unknown E-CommereShippingData · training data not disclosed in enough detail

Unknown Fitness_Club · training data not disclosed in enough detail

Unknown Food_Delivery_Time · training data not disclosed in enough detail

Unknown GiveMeSomeCredit · training data not disclosed in enough detail

Unknown hazelnut-spread-contaminant-detection · training data not disclosed in enough detail

Unknown healthcare_insurance_expenses · training data not disclosed in enough detail

Seen, split unclear heloc · the dataset was used; the source does not say which part

Evidence (2)
“Diabetes130US heloc”
arxiv.org · paper · Appendix B, Table B.1 (OpenML dataset ID 45026) · retrieved 2026-10-15
“In TabArena-v0.1, TabDPT is the only model directly pretrained on benchmark datasets, covering seven tasks (KDDCup09_appetency, QSARTID-11, Amazon_employee_access, APSFailure, wine_quality, Diabetes130US, heloc).”
arxiv.org · paper · Appendix E.6 · retrieved 2026-10-15

Unknown hiva_agnostic · training data not disclosed in enough detail

Unknown houses · training data not disclosed in enough detail

Unknown HR_Analytics_Job_Change_of_Data_Scientists · training data not disclosed in enough detail

Unknown in_vehicle_coupon_recommendation · training data not disclosed in enough detail

Unknown Is-this-a-good-customer · training data not disclosed in enough detail

Seen, split unclear kddcup09_appetency · the dataset was used; the source does not say which part

Evidence (2)
“KDDCup09_appetency”
arxiv.org · paper · Appendix B, Table B.1 (OpenML dataset ID 1111) · retrieved 2026-10-15
“In TabArena-v0.1, TabDPT is the only model directly pretrained on benchmark datasets, covering seven tasks (KDDCup09_appetency, QSARTID-11, Amazon_employee_access, APSFailure, wine_quality, Diabetes130US, heloc).”
arxiv.org · paper · Appendix E.6 · retrieved 2026-10-15

Unknown Marketing_Campaign · training data not disclosed in enough detail

Unknown maternal_health_risk · training data not disclosed in enough detail

Unknown miami_housing · training data not disclosed in enough detail

Unknown MIC · training data not disclosed in enough detail

Unknown NATICUSdroid · training data not disclosed in enough detail

Unknown online_shoppers_intention · training data not disclosed in enough detail

Unknown physiochemical_protein · training data not disclosed in enough detail

Unknown polish_companies_bankruptcy · training data not disclosed in enough detail

Unknown qsar-biodeg · training data not disclosed in enough detail

Seen, split unclear QSAR-TID-11 · the dataset was used; the source does not say which part

Evidence (2)
“QSAR-TID-11”
arxiv.org · paper · Appendix B, Table B.1 (OpenML dataset ID 3050) · retrieved 2026-10-15
“In TabArena-v0.1, TabDPT is the only model directly pretrained on benchmark datasets, covering seven tasks (KDDCup09_appetency, QSARTID-11, Amazon_employee_access, APSFailure, wine_quality, Diabetes130US, heloc).”
arxiv.org · paper · Appendix E.6 · retrieved 2026-10-15

Unknown QSAR_fish_toxicity · training data not disclosed in enough detail

Unknown SDSS17 · training data not disclosed in enough detail

Unknown seismic-bumps · training data not disclosed in enough detail

Unknown splice · training data not disclosed in enough detail

Unknown students_dropout_and_academic_success · training data not disclosed in enough detail

Unknown superconductivity · training data not disclosed in enough detail

Unknown taiwanese_bankruptcy_prediction · training data not disclosed in enough detail

Unknown website_phishing · training data not disclosed in enough detail

Seen, split unclear wine_quality · the dataset was used; the source does not say which part

Evidence (2)
“wine_quality”
arxiv.org · paper · Appendix B, Table B.1 (OpenML dataset ID 44136) · retrieved 2026-10-15
“In TabArena-v0.1, TabDPT is the only model directly pretrained on benchmark datasets, covering seven tasks (KDDCup09_appetency, QSARTID-11, Amazon_employee_access, APSFailure, wine_quality, Diabetes130US, heloc).”
arxiv.org · paper · Appendix E.6 · retrieved 2026-10-15

Unknown jm1 · training data not disclosed in enough detail

Notes

Pre-trained only on real OpenML data (self-supervised targets), no synthetic data. The paper deduplicated training data against its own evaluation suites CC18 and CTR23 ('Any dataset pairs with unusual similarities were manually evaluated and those found to be related were removed from training.'), not against TabArena. Seven training datasets overlap TabArena-v0.1 per TabArena's own analysis. The paper does not say whether v1.0 and v1.1 used the same data; the TabDPT-Turbo paper reports 112 datasets for v1.1 vs 123 in Table B.1 (see open questions).

Sources