TabICLv2: A better, faster, scalable, and open tabular foundation model.arXiv preprint arXiv:2602.11139

representative citing papers

citing papers explorer

Showing 41 of 41 citing papers.

• STRABLE: Benchmarking Tabular Machine Learning with Strings cs.LG · 2026-05-12 · unverdicted · none · ref 49

  A new corpus of 108 mixed string-numeric tables shows that advanced tabular learners with basic string embeddings perform well on most real-world data, while large LLM encoders help on free-text heavy tables.

• Beyond IID: How General Are Tabular Foundation Models, Really? cs.LG · 2026-06-29 · unverdicted · none · ref 197

  Tabular foundation models excel on tiny- to medium-sized IID data but are outperformed by traditional tree-based and deep learning models on non-IID, large, and high-dimensional datasets, based on evaluations across 11 models and 142 datasets in the new BeyondArena benchmark.

• FlexTab: A Flexible Encoder-Decoder Architecture for In-Context Learning Across Diverse Tabular Tasks cs.LG · 2026-06-29 · unverdicted · none · ref 52 · 2 links

  FlexTab shows a shared encoder with task-specific decoders trained on unlabeled tables can achieve SOTA on classification, regression, anomaly detection and entity matching while staying competitive on relational entity classification.

• Computational Identifiability cs.LG · 2026-06-08 · unverdicted · none · ref 38

  The paper defines computational identifiability as success of a finite search procedure in finding an empirical estimator for a causal query within error tolerance, conditional on the search assumptions and procedure.

• TS-ICL: A Flexible Time-Indexed Foundation Model for Time Series via In-Context Learning cs.LG · 2026-06-04 · unverdicted · none · ref 35

  TS-ICL introduces a probabilistic in-context learning encoder-regressor Transformer that unifies forecasting and imputation for time series via timestamp-aligned regression trained on synthetic causal data.

• Speedrunning Tabular Foundation Model Pretraining cs.LG · 2026-06-02 · unverdicted · none · ref 7

  A speedrun benchmark for nanoTabPFN pretraining reports a record of 0.92 minutes to target performance, an 81x speedup over the 74.32-minute baseline using 22x fewer synthetic datasets.

• TabPrep: Closing the Feature Engineering Gap in Tabular Benchmarks cs.LG · 2026-06-01 · unverdicted · none · ref 2

  TabPrep is a new feature engineering pipeline that targets three data patterns and improves performance of tree-based, neural, linear, and foundation models on tabular benchmarks, often more than model architecture changes.

• CalArena: A Large-Scale Post-Hoc Calibration Benchmark cs.LG · 2026-05-28 · conditional · none · ref 64

  CalArena is a large-scale benchmark that evaluates dozens of post-hoc calibration methods using Post-Hoc Improvement (PHI) in proper scoring rules and finds that smooth functions outperform binning while dedicated multiclass methods are required in high-dimensional settings.

• SurvivalPFN: Amortizing Survival Prediction via In-Context Bayesian Inference cs.LG · 2026-05-15 · unverdicted · none · ref 84

  SurvivalPFN amortizes Bayesian survival analysis for right-censored data by pretraining a prior-data fitted network on synthetic identifiable DGPs and then performing in-context inference, achieving competitive results on 61 real datasets.

• MulTaBench: Benchmarking Multimodal Tabular Learning with Text and Image cs.LG · 2026-05-11 · unverdicted · none · ref 83

  MulTaBench is a new collection of 40 image-tabular and text-tabular datasets designed to test target-aware representation tuning in multimodal tabular models.

• PFN-TS: Thompson Sampling for Contextual Bandits via Prior-Data Fitted Networks stat.ML · 2026-05-11 · unverdicted · none · ref 23

  PFN-TS converts PFN posterior predictives into mean-reward samples for Thompson sampling using a subsampled predictive CLT, with consistency proofs, regret bounds, and strong empirical performance on synthetic and real bandit benchmarks.

• Data Language Models: A New Foundation Model Class for Tabular Data cs.AI · 2026-05-07 · unverdicted · none · ref 14

  Schema-1 is the first Data Language Model that natively understands raw tabular data and outperforms gradient-boosted ensembles, AutoML, and prior tabular foundation models on row-level prediction and imputation tasks.

• TFM-Retouche: A Lightweight Input-Space Adapter for Tabular Foundation Models cs.LG · 2026-05-07 · unverdicted · none · ref 24 · 2 links

  TFM-Retouche is an architecture-agnostic input-space residual adapter that improves tabular foundation model accuracy on 51 datasets by learning input corrections through the frozen backbone, with an identity guard to fall back to the original model.

• RamanBench: A Large-Scale Benchmark for Machine Learning on Raman Spectroscopy cs.LG · 2026-05-03 · unverdicted · none · ref 48

  RamanBench unifies 74 datasets into the first large-scale reproducible benchmark for ML on Raman spectra, finding tabular foundation models outperform baselines but no method generalizes across datasets.

• TimEE: End-to-end Time Series Classification via In-Context Learning cs.LG · 2026-07-08 · conditional · none · ref 32

  A 4.5M-parameter transformer meta-trained on synthetic VARX-generated classification tasks achieves state-of-the-art ROC AUC on the UCR time series classification benchmark via in-context learning with no per-dataset training.

• TabPATE: Differentially Private Tabular In-Context Learning Without Public Data cs.LG · 2026-06-30 · unverdicted · none · ref 7

  TabPATE applies a PATE-style private aggregation to synthetic tabular queries generated from feature ranges, enabling private in-context learning with near-random membership inference success while keeping competitive utility.

• Efficient Adaptive Data Acquisition via Pretrained Belief Representations cs.LG · 2026-06-23 · unverdicted · none · ref 52

  POLAR uses pretrained predictive foundation models as fixed belief-state encoders and trains only a lightweight policy head on top for amortised Bayesian experimental design, optimisation, and active learning.

• Bounded Context Management for Tabular Foundation Models on Stream Learning cs.LG · 2026-06-17 · conditional · none · ref 10

  CURE is a context management policy using entropy-gated admission and redundancy-aware eviction that delivers up to 27% relative improvement over classical stream learners on seven streams while remaining robust across TFM backbones.

• CRUMB: Efficient Prior Fitted Network Inference via Distributionally Matched Context Batching cs.LG · 2026-06-09 · unverdicted · none · ref 16

  CRUMB speeds up PFN inference on large tabular datasets by clustering queries and selecting MMD-matched context subsets, outperforming prior selection methods on the 51-dataset TabArena benchmark across three architectures while handling covariate drift.

• In-Context Learning for Latent Space Bayesian Optimization cs.LG · 2026-06-08 · unverdicted · none · ref 24

  Complementing tabular foundation model pretraining with LSBO-specific synthetic tasks and a regularizer yields strong performance on held-out molecular optimization benchmarks.

• LUCoS: Latent Unsupervised Context Selection for Tabular Foundation Models cs.LG · 2026-05-26 · unverdicted · none · ref 7

  LUCoS replaces raw tabular geometry with unsupervised PFN latent embeddings for medoid-based context selection and ranks first on mean AUC, ACC, and F1 across 67 datasets and six budgets.

• Shaping the Prior: How Synthetic Task Distributions Determine Tabular Foundation Model Quality cs.LG · 2026-05-18 · unverdicted · none · ref 23

  O'Prior, a compositional synthetic prior with hierarchical SCMs, realism engines, stress modules, and curriculum protocols, improves tabular foundation model accuracy and robustness on real benchmarks when architecture and compute are held fixed.

• Pocket Foundation Models: Distilling TFMs into CPU-Ready Gradient-Boosted Trees cs.LG · 2026-05-18 · accept · none · ref 1

  Distilling TabICLv2 into XGBoost via stratified OOF labeling yields 0.882 macro-mean AUC (96.5% of teacher) at 1.9 ms CPU across 153 datasets, with significant gains over tuned CatBoost on low-dimensional data.

• Causal Foundation Models with Continuous Treatments cs.LG · 2026-05-14 · conditional · none · ref 40

  A transformer meta-trained on a novel continuous-treatment data-generating prior reconstructs individual treatment-response curves from observational data via in-context learning and reports SOTA versus task-specific causal models.

• KGPFN: Unlocking the Potential of Knowledge Graph Foundation Model via In-Context Learning cs.AI · 2026-05-14 · unverdicted · none · ref 34

  KGPFN pretrains on multiple KGs to learn relation patterns, then performs query-specific reasoning by encoding local context with NBFNet and global context via retrieved instances aggregated in a PFN with feature- and sample-level attention.

• Online Sharp-Calibrated Bayesian Optimization cs.LG · 2026-05-11 · unverdicted · none · ref 27

  OSCBO adaptively balances Gaussian process sharpness and calibration in Bayesian optimization by casting hyperparameter selection as constrained online learning, while preserving sublinear regret bounds.

• In-Context Black-Box Optimization with Unreliable Feedback cs.LG · 2026-05-07 · unverdicted · none · ref 25

  FICBO pretrains a feedback-aware transformer with a structured prior on feedback distortion to adaptively exploit or ignore unreliable auxiliary signals during in-context black-box optimization.

• Tabular foundation models for in-context prediction of molecular properties cs.LG · 2026-04-17 · unverdicted · none · ref 15

  Tabular foundation models achieve high accuracy in molecular property prediction through in-context learning, with up to 100% win rates on MoleculeACE tasks when paired with CheMeleon embeddings.

• Benchmarking Optimizers for MLPs in Tabular Deep Learning cs.LG · 2026-04-16 · unverdicted · none · ref 10

  Muon optimizer outperforms AdamW across 17 tabular datasets when training MLPs under a shared protocol.

• KumoRFM-2: Scaling Foundation Models for Relational Learning cs.LG · 2026-04-14 · unverdicted · none · ref 12

  KumoRFM-2 pre-trains on synthetic and real relational data across row, column, foreign-key and cross-sample axes, injects task information early, and achieves up to 8% gains over supervised baselines on 41 benchmarks in few-shot and fine-tuned regimes while handling billion-scale datasets.

• Exploring Differences Between Tabular Enterprise Data and Public Benchmarks cs.LG · 2026-06-29 · unverdicted · none · ref 8

  Enterprise tabular data differs from public benchmarks in ways that prevent good generalization of models like TabPFN, TabICL, and ConTextTab between the two domains.

• Are Tabular Foundation Models Robust to Realistic Query Distribution Shifts in Microbiome Data? cs.LG · 2026-06-23 · unverdicted · none · ref 20

  Tabular foundation models for gut microbiome data show degraded performance under perturbations like zero-imputation and sparsification even when discriminative taxa are retained, with zero-imputation most harmful and greater sensitivity than random forests.

• Rejections Based on Predictive Uncertainty Enable Reliable Routine Soil Spectroscopy cs.LG · 2026-06-19 · unverdicted · none · ref 49

  Reject-to-remeasure combines probabilistic ML on soil spectra with uncertainty-guided rejection to meet user accuracy targets on a Québec dataset while lowering measurement costs.

• Simultaneous hyperkinetic movement disorders phenotyping: a cross-cohort pediatric transfer study using routine videos, markerless pose estimation and a tabular foundation model cs.CV · 2026-06-04 · unverdicted · none · ref 9

  A pose-estimation plus tabular foundation model pipeline trained on 25 adults transfers to 12 pediatric hyperkinetic movement disorder cases with lightweight final-layer calibration, raising Hamming accuracy from 0.804 to 0.839 and Jaccard index from 0.548 to 0.633 on held-out patients.

• When Tabular Foundation Models Transfer Across Modalities: A Systematic Evaluation Across 95 Datasets, 7 Modalities, and Two Regimes cs.LG · 2026-06-01 · unverdicted · none · ref 8

  A tabular foundation model pipeline with ETF preprocessing transfers across 7 modalities on 95 datasets, matching lightweight tuned baselines on frozen features at much higher speed while providing calibration for deployment.

• Modular Multimodal Classification Without Fine-Tuning: A Simple Compositional Approach cs.LG · 2026-05-20 · unverdicted · none · ref 23

  CoMET achieves strong multimodal classification performance by composing frozen modality encoders, PCA compression, and tabular foundation models without any training, reaching state-of-the-art on diverse benchmarks including large-scale hierarchical tasks.

• Distilling Tabular Foundation Models for Structured Health Data cs.LG · 2026-05-18 · unverdicted · none · ref 7

  Leakage-aware distillation transfers at least 90% of tabular foundation model AUC to lightweight students across 19 health datasets, with 26x CPU speedup and preserved calibration/fairness.

• Tabular Foundation Model for Generative Modelling cs.LG · 2026-05-10 · unverdicted · none · ref 67

  TabFORGE generates high-quality synthetic tabular data by leveraging pretrained causality-aware representations in a two-stage diffusion-decoder architecture that mitigates latent distribution shifts.

• TabCF: Distributional Control Function Estimation with Tabular Foundation Models stat.ML · 2026-05-07 · unverdicted · none · ref 34

  TabCF is a tuning-light method using tabular foundation models for control function regression to estimate distributional causal effects such as interventional means and quantiles.

• The Good, the Bad, and the Ugly of Markov Boundary for Tabular Prediction cs.LG · 2026-05-28 · unreviewed · ref 34
• TabH2O: A Unified Foundation Model for Tabular Prediction cs.LG · 2026-05-18 · unreviewed · ref 18