Tabular / Model
TabICL
- Organisation
- Inria (SODA team)
- Family
- TabICL · version v1
- Released
- 2025-02-08 (arXiv v1 submission date https://arxiv.org/abs/2502.05564)
- Checkpoints
- jingang/TabICL: tabicl-classifier-v1-20250208.ckpt, jingang/TabICL-clf: tabicl-classifier-v1-0208.ckpt
- Weights
- Open · BSD-3-Clause
- Links
- PaperModel cardCode
Training data disclosure: Full list
Level determined for 100% of benchmark cells.
- TestTest data seen0
- SeenSeen, split unclear0
- TrainTraining data seen0
- Related data seen0
- NoNot seen51
- ?Unknown0
“TabICL is pre-trained exclusively on synthetic datasets.”
Other declared training data
-
TabICL synthetic prior (SCM and tree-based generators) · split: unknown
Evidence
“TabICL is pre-trained exclusively on synthetic datasets.”
arxiv.org · paper · Section 3.4 (prior) · retrieved 2026-10-15
Benchmark cells
TabArena
Not seen airfoil_self_noise · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Amazon_employee_access · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen anneal · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Another-Dataset-on-used-Fiat-500 · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen APSFailure · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen bank-marketing · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Bank_Customer_Churn · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Bioresponse · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen blood-transfusion-service-center · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen churn · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen coil2000_insurance_policies · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen concrete_compressive_strength · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen credit-g · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen credit_card_clients_default · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen customer_satisfaction_in_airline · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen diabetes · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Diabetes130US · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen diamonds · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen E-CommereShippingData · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Fitness_Club · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Food_Delivery_Time · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen GiveMeSomeCredit · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen hazelnut-spread-contaminant-detection · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen healthcare_insurance_expenses · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen heloc · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen hiva_agnostic · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen houses · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen HR_Analytics_Job_Change_of_Data_Scientists · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen in_vehicle_coupon_recommendation · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Is-this-a-good-customer · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen kddcup09_appetency · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen Marketing_Campaign · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen maternal_health_risk · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen miami_housing · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen MIC · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen NATICUSdroid · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen online_shoppers_intention · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen physiochemical_protein · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen polish_companies_bankruptcy · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen qsar-biodeg · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen QSAR-TID-11 · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen QSAR_fish_toxicity · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen SDSS17 · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen seismic-bumps · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen splice · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen students_dropout_and_academic_success · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen superconductivity · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen taiwanese_bankruptcy_prediction · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen website_phishing · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Not seen wine_quality · absent from the complete list of training data
Evidence (1)
“TabICL is pre-trained exclusively on synthetic datasets.”
Notes
Classification only.
Sources
- TabICL: A Tabular Foundation Model for In-Context Learning on Large Data (arXiv v2) · paper · retrieved 2026-10-15
- soda-inria/tabicl README · readme · retrieved 2026-10-15
- jingang/TabICL model card · model card · retrieved 2026-10-15