# TabPFN

7 related articles

Subscribe to updates

arXiv Nov 11, 2026
A frozen tabular foundation model as candidate generator: propeller-compatible stern designs rise from 1.4% to 99.3%
Engineering design often has only 10 to 1,000 evaluated designs because simulations and experiments are costly. GenPFN trains no new generator: a frozen TabPFN classifier reads labeled designs, candidates are interpolated between desired designs and updated with classifier gradients and Gaussian noise, and every candidate is evaluated by the problem's own constraints and objectives. Across fourteen engineering problems, GenPFN v1 has higher class satisfaction and lower realism distance than the tested CTGAN, TVAE and TabDDPM; in the stern case, 993 of 1,000 candidates passed geometric checks and met the propeller-placement area requirement, versus 28 of 2,000 random context designs.
Engineering design often has only 10 to 1,000 evaluated designs because simulations and experiments are costly. GenPFN trains no new generator: a frozen TabPFN classifier reads labeled designs, candidates are interpolated between desired designs and updated with classifier gradients and Gaussian noise, and every candidate is evaluated by the problem's own constraints and objectives. Across fourteen engineering problems, GenPFN v1 has higher class satisfaction and lower realism distance than the tested CTGAN, TVAE and TabDDPM; in the stern case, 993 of 1,000 candidates passed geometric checks and met the propeller-placement area requirement, versus 28 of 2,000 random context designs.
Noah J. Bagazinski（First author, Corresponding author）
GenPFN TabPFN Data Augmentation In-Context Learning
bioRxiv Nov 9, 2026
TabPFN 3.5 matches species distribution models without tuning, and fine-tuning makes it the benchmark's best single model
Predicting where a species occurs from presence records alone has long relied on models such as MaxEnt that must be tuned for each dataset. On a standard benchmark of 225 species in six regions, TabPFN 3.5 reached a mean AUC of 0.732 without tuning, matching the best published five-model ensemble; fine-tuned on presence–background tasks from other regions it rose to 0.736, the best single model on the benchmark and significantly above MaxEnt's 0.728. The authors caution that six regions cannot establish the advantage in new regions, and that the model is sensitive to environmentally biased records.
Predicting where a species occurs from presence records alone has long relied on models such as MaxEnt that must be tuned for each dataset. On a standard benchmark of 225 species in six regions, TabPFN 3.5 reached a mean AUC of 0.732 without tuning, matching the best published five-model ensemble; fine-tuned on presence–background tasks from other regions it rose to 0.736, the best single model on the benchmark and significantly above MaxEnt's 0.728. The authors caution that six regions cannot establish the advantage in new regions, and that the model is sensitive to environmentally biased records.
Tokyo Metropolitan University Kyoto University Fujitsu (Japan)
Hideyuki Doi（First author, Corresponding author）
Tabular Foundation Models TabPFN Fine-tuning
arXiv Nov 7, 2026
Cancer outcome models stop splitting by ancestry: DA-ICL models the genetic continuum and tops mean AUROC in 10 of 18 TCGA tasks
Omics-based cancer outcome prediction usually assigns samples to discrete ancestry groups and fits a separate model within each group, yet accuracy falls continuously with genetic distance from the training set. DA-ICL treats ancestry as a continuous variable, projects genotypes into a low-dimensional space with principal component analysis, builds a context from nearby samples at each ancestry point, and lets a pretrained tabular foundation model predict at inference time, with distribution-aware context enhancement adding distant but distributionally similar samples. Across 18 tasks spanning three TCGA cancer types, three omics modalities and two clinical endpoints, DA-ICL reaches the highest mean AUROC on 10 tasks for African-ancestry samples, with gains that vary by task.
Omics-based cancer outcome prediction usually assigns samples to discrete ancestry groups and fits a separate model within each group, yet accuracy falls continuously with genetic distance from the training set. DA-ICL treats ancestry as a continuous variable, projects genotypes into a low-dimensional space with principal component analysis, builds a context from nearby samples at each ancestry point, and lets a pretrained tabular foundation model predict at inference time, with distribution-aware context enhancement adding distant but distributionally similar samples. Across 18 tasks spanning three TCGA cancer types, three omics modalities and two clinical endpoints, DA-ICL reaches the highest mean AUROC on 10 tasks for African-ancestry samples, with gains that vary by task.
Weizhi Li（First author, Corresponding author）Yan Cui（Corresponding author）
In-Context Learning TabPFN Cancer Outcome Prediction Distribution-Aware Context Enhancement
arXiv Nov 6, 2026
SRAM design stops fixing the bitcell first: OpenYieldV2 optimizes 6T and 10T together, raising hypervolume by up to 17.3%
The open-source SRAM design tool OpenYieldV2 turns bitcell topology from a fixed input into an optimizable variable searched jointly with transistor sizing and threshold-voltage assignment. Previously OpenRAM generated layouts without optimizing, and OpenYield's high-fidelity optimization covered only a fixed 6T cell. OpenYieldV2 compiles 6T and 10T macros under identical high-fidelity conditions, predicts performance with a few-shot TabPFN surrogate, and searches with a coarse-to-refined multi-objective optimizer; in mixed-topology scenarios its hypervolume improves by up to 17.3% over the strongest single-topology baseline.
The open-source SRAM design tool OpenYieldV2 turns bitcell topology from a fixed input into an optimizable variable searched jointly with transistor sizing and threshold-voltage assignment. Previously OpenRAM generated layouts without optimizing, and OpenYield's high-fidelity optimization covered only a fixed 6T cell. OpenYieldV2 compiles 6T and 10T macros under identical high-fidelity conditions, predicts performance with a few-shot TabPFN surrogate, and searches with a coarse-to-refined multi-objective optimizer; in mixed-topology scenarios its hypervolume improves by up to 17.3% over the strongest single-topology baseline.
Can Wang（First author）
OpenYieldV2 Joint Topology-Sizing Optimization TabPFN Multi-Objective Optimization
arXiv Sep 26, 2026
M²PFN turns a frozen TabPFN into a multimodal Alzheimer's predictor, reaching 65.55% macro-F1 on ADNI and transferring to two external cohorts without retraining
The work proposes M²PFN, an end-to-end framework that back-propagates task gradients into 3D-MRI and tabular encoders through differentiable inference, aligns the two modalities into a shared subspace matched to the in-context learning prior via disentanglement and a contrastive objective, and folds in a frozen tabular-only prediction through a learnable gated shortcut; on ADNI (n=2240, three-class CN/MCI/AD) it attains 65.55% macro-F1 and 82.21% macro-AUC, surpassing the compared unimodal and multimodal baselines, regresses baseline MMSE by swapping only the head (1250-subject sub-cohort, test MAE 1.743), and achieves the best AUC and lowest MMSE MAE across all baselines on two external cohorts (OASIS-3 and SCAN) with no retraining, transferring even when the cognitive instrument changes.
The work proposes M²PFN, an end-to-end framework that back-propagates task gradients into 3D-MRI and tabular encoders through differentiable inference, aligns the two modalities into a shared subspace matched to the in-context learning prior via disentanglement and a contrastive objective, and folds in a frozen tabular-only prediction through a learnable gated shortcut; on ADNI (n=2240, three-class CN/MCI/AD) it attains 65.55% macro-F1 and 82.21% macro-AUC, surpassing the compared unimodal and multimodal baselines, regresses baseline MMSE by swapping only the head (1250-subject sub-cohort, test MAE 1.743), and achieves the best AUC and lowest MMSE MAE across all baselines on two external cohorts (OASIS-3 and SCAN) with no retraining, transferring even when the cognitive instrument changes.
University of Southern California
Lujia Zhong（First author）
M²PFN TabPFN In-Context Learning Alzheimer's disease
bioRxiv Sep 18, 2026
Data-driven predictive design of engineered living hydrogels
Using the tabular foundation model TabPFN informed by a small library of living hydrogels, this work predicts macroscopic material properties of Escherichia coli-produced living hydrogels containing CsgA-based fibres fused to genetically encoded PEG-like biopolymers from genetic and process parameters, achieving the strongest prediction for storage modulus G' (R2 = 85.1%) on an independent validation set with a 48.0% RMSE reduction versus linear regression, and further enabling property-guided design to identify parameters for desired properties.
Using the tabular foundation model TabPFN informed by a small library of living hydrogels, this work predicts macroscopic material properties of Escherichia coli-produced living hydrogels containing CsgA-based fibres fused to genetically encoded PEG-like biopolymers from genetic and process parameters, achieving the strongest prediction for storage modulus G' (R2 = 85.1%) on an independent validation set with a 48.0% RMSE reduction versus linear regression, and further enabling property-guided design to identify parameters for desired properties.
Saarland University
Geisler Muñoz-Guamuro（First author）Wilfried Weber（Corresponding author）
TabPFN Tabular Foundation Model Living hydrogels Engineered living materials
Journal of Chemical Information and Modeling Sep 14, 2026
In-Context Learning Meets Small Molecule Property Prediction: Benchmarking Novel Machine Learning Approaches
This study comprehensively benchmarked tabular foundation models (TFMs) based on in-context learning for small organic molecule property prediction, comparing several TFMs with multiple machine learning methods across 11 data sets (regression, random and structure-aware splits, up to 10,000 molecules each), and found that TFMs consistently outperform XGBoost, CatBoost, multilayer perceptrons, and other descriptor-based methods even with careful hyperparameter selection for the baselines, achieve accuracy on par with or better than graph-based methods including those pretrained on chemical data, with Uni-Mol2 slightly outperforming TFMs in some experiments, while retrieval (selecting the 500 closest neighbors by Tanimoto similarity for each test molecule) yields further improvement on relat
This study comprehensively benchmarked tabular foundation models (TFMs) based on in-context learning for small organic molecule property prediction, comparing several TFMs with multiple machine learning methods across 11 data sets (regression, random and structure-aware splits, up to 10,000 molecules each), and found that TFMs consistently outperform XGBoost, CatBoost, multilayer perceptrons, and other descriptor-based methods even with careful hyperparameter selection for the baselines, achieve accuracy on par with or better than graph-based methods including those pretrained on chemical data, with Uni-Mol2 slightly outperforming TFMs in some experiments, while retrieval (selecting the 500 closest neighbors by Tanimoto similarity for each test molecule) yields further improvement on relat
Russian Academy of Sciences
Dmitriy D. Matyushin（First author）
Tabular Foundation Models TabPFN In-Context Learning Molecular Property Prediction Retrieval-Augmented Inference

You’re all caught up

A frozen tabular foundation model as candidate generator: propeller-compatible stern designs rise from 1.4% to 99.3%

TabPFN 3.5 matches species distribution models without tuning, and fine-tuning makes it the benchmark's best single model

Cancer outcome models stop splitting by ancestry: DA-ICL models the genetic continuum and tops mean AUROC in 10 of 18 TCGA tasks

SRAM design stops fixing the bitcell first: OpenYieldV2 optimizes 6T and 10T together, raising hypervolume by up to 17.3%

M²PFN turns a frozen TabPFN into a multimodal Alzheimer's predictor, reaching 65.55% macro-F1 on ADNI and transferring to two external cohorts without retraining

Data-driven predictive design of engineered living hydrogels

In-Context Learning Meets Small Molecule Property Prediction: Benchmarking Novel Machine Learning Approaches