Prior-Data Fitted Networks

Also known as PFN

PFNs are models, often transformers, trained on many synthetic datasets so they can make predictions for a new dataset from its examples, without task-specific optimization. Their synthetic priors encode possible data-generating processes, and researchers adapt them for tabular prediction, time-series forecasting, survival analysis, causal inference, graph learning, and Bayesian optimization. Common approaches include amortizing Bayesian posterior prediction, designing structured or hybrid priors, and adapting pretrained representations or inference procedures to new tasks. A central challenge is that performance and uncertainty depend on how well the training prior matches deployment data: models can be fast and effective in familiar settings yet retain errors on rare or misspecified cases.

Latest papers

48 papers14 papers in the last four weeks, against 1 the four weeks before. 0.1% of all new papers.

GAS-PFN: Graph-Aware Spatiotemporal Prior-Data Fitted Networks for Zero-Shot Traffic Forecasting
Oct 27, 2026Joseph Natter, Long Vu, Yifan Zhang +3Traffic Forecasting Spatiotemporal Forecasting
Kennesaw State University · Missouri State University
Prior-Data Fitted Networks for In-Context Off-Policy Evaluation
Oct 25, 2026Shuguang Yu, Ting Li, Fan ZhouOff-Policy Evaluation Prior-Data Fitted Networks
The Hong Kong Polytechnic University, Hong Kong, China · Shanghai University of Finance and Economics, Shanghai, China
Debiased Prior-Data Fitted Networks for Amortized Bayesian Semi-Parametric Inference
Oct 20, 2026Valentyn Melnychuk, Emil Javurek, Stefan FeuerriegelPrior-Data Fitted Networks Amortized Inference
LMU Munich & Munich Center for Machine Learning, Munich, Germany
Path to Prior-Data Fitted Relational Models via Learned Latent Multi-table Representations
Oct 16, 2026Mohammad Sadeq Abolhasani, Viswanath GanapathyRelational Foundation Models Synthetic Data Pretraining
SAP Labs
SelectPFN: Amortized Bayesian Variable Selection for Sparse Nonlinear Regression with Prior-Data Fitted Networks
Oct 15, 2026Bin LuoFeature Selection Prior-Data Fitted Networks
School of Data Science and Analytics, Kennesaw State University
Characterizing the Separability Gap in Prior-Data Fitted Networks
Oct 10, 2026Julius Kobialka, Manuel Hülskamp, Thomas Nagler +1Prior-Data Fitted Networks Bayesian Inference
Department of Statistics, LMU Munich · Munich Center for Machine Learning
How Well Do Prior-Data Fitted Networks Amortize Bayesian Inference?
Oct 9, 2026Manuel Hülskamp, Julius Kobialka, Emanuel Sommer +2Bayesian Model Averaging Prior-Data Fitted Networks
LMU Munich and Munich Center for Machine Learning
PDE-PFN: Prior-Data Fitted Neural PDE Solver
Oct 9, 2026Jaehyeon Park, Mingu Kang, Dongseok Lee +5Neural PDE Solvers PDE Solving
SolverX · KAIST · Seoul National University +3
Amortized Off-Policy Evaluation for LLMs
Oct 7, 2026Younwoo Choi, Leo Feng, Vincent Liu +1Off-Policy Evaluation LLM Evaluation
RBC Borealis · University of Toronto · Vector Institute
Continual Learning without Continual Training
Oct 7, 2026Nikita Narayanan, Ritham Majumdarr, Sonali ParbhooContinual Learning Class-Incremental Learning
Imperial College London
ProximalFM: Amortized Proximal Causal Inference under Hidden Confounding
Oct 6, 2026Christophe Muller, Ayub Kharel, Alex Luedtke +6Causal Inference Amortized Inference
University of Oxford · Harvard University · University of Illinois Urbana-Champaign +5
Adaptive Mean Estimation by In-Context Learning: A Gradient-Flow Analysis
Oct 6, 2026Martin Eppert, Krishna Balasubramanian, Subhro Ghosh +2In-Context Learning Prior-Data Fitted Networks
National University of Singapore · University of California, Davis · Princeton University
Adapting prior-data fitted networks for tabular anomaly detection
Oct 5, 2026Maximilian Bershtman, Niv CohenTabular Anomaly Detection Unsupervised Anomaly Detection
Faculty of Electrical and Computer Engineering Technion – Israel Institute of Technology Haifa, Israel
LinearPFN: Amortized Variable Selection for Linear Models with Interactions
Oct 5, 2026Louis Schiekiera, Max Zimmer, Christophe Roux +3Model Selection Prior-Data Fitted Networks
Institute of Psychology, Humboldt-Universität zu Berlin, Berlin, Germany · Department of Education and Psychology, Freie Universität Berlin, Berlin, Germany · Department for AI in Society, Science, and Technology, Zuse Institute Berlin, Germany +1
IQS-BO: In-Context Query Selection for Bayesian Optimisation
Oct 1, 2026Luca Geminiani, Nadja KleinBayesian Optimization Prior-Data Fitted Networks
Scientific Computing Center (SCC), Karlsruhe Institute of Technology (KIT)
When and Why LLM Causal Priors Help: Closed-Loop Prior Selection for Amortized Causal Inference
Sep 7, 2026Haohao ZhouCausal Inference Causal Effect Estimation
National Key Laboratory of Information Systems Engineering National University of Defense Technology, Changsha, Hunan, China
RecPFN: Prior-Fitted Networks for In-Context-Based Recommendations
Aug 20, 2026En Zhi Tan, Jia Xiang Lim, Bryan Lijie Chew +2Recommender Systems Sequential Recommendation
SAP SE Singapore
Curriculum Matters: Data-Efficient Relational PFN Pretraining with Synthetic Data
Jul 31, 2026Mohammad Sadeq Abolhasani, Viswanath GanapathySynthetic Data Pretraining Synthetic Data Generation
SAP Labs, LLC. Palo Alto, CA, United States
PerturbPFN: Probing the Limits of Synthetic Priors in Drug Perturbation Modelling
Jul 26, 2026Yuche Gao, José Miguel Hernández-Lobato, Siyuan GuoSynthetic Data Pretraining Structural Causal Models
University of Cambridge, Cambridge, United Kingdom · Prior Labs, Freiburg, Germany
TabPFN beyond Tabular Data: Calibration and Accuracy on Multimodal Embeddings
Jul 13, 2026Jingxiang Zhang, Lujia Zhong, Zijie Zhu +2Multimodal Embedding Multimodal Classification
University of Southern California
A Fair Evaluation of Graph Foundation Models for Node Property Prediction
Jun 23, 2026Oleg Platonov, Gleb Bazhenov, Dmitry Eremeev +1Graph Neural Networks Graph Foundation Models
1HSE University · 2Yandex Research
A Causal DAG Prior for Synthetic Time-Series Classification Datasets
Jun 19, 2026Franco Martino O'Rourke, Ana Trisovic, Dimitris BertsimasStructural Causal Models Multivariate Time Series Classification
Massachusetts Institute of Technology, Cambridge, MA, USA.
DCD-PFN: A Decoupling-Aware Foundation Model for Causal Discovery
Jun 19, 2026Zhengkang Guan, Yikang Chen, Yi He +5Causal Foundation Models Structural Causal Models
College of Computer Science and Technology Zhejiang University
Temporal Causal Prior-Data Fitted Networks for Panel Data with Learned Reliability Signals
Jun 18, 2026Shravan Talupula, Saurabh SharmaCausal Foundation Models Causal Effect Estimation
ProfitOps Inc.
CRUMB: Efficient Prior Fitted Network Inference via Distributionally Matched Context Batching
Jun 9, 2026Jamie Heredge, Mattia J. Villani, Pranav Deshpande +2In-Context Learning Prior-Data Fitted Networks
Global Technology Applied Research, JPMorganChase, New York, NY 10001, USA
$α$ -PFN: Fast Entropy Search via In-Context Learning
Jun 5, 2026Herilalaina Rakotoarison, Steven Adriaensen, Tom Viering +4Bayesian Optimization Prior-Data Fitted Networks
University of Helsinki · University of Freiburg · Delft University of Technology +2
Causal Longitudinal Prior-Fitted Networks for Counterfactual Outcome Prediction
Jun 4, 2026Amirhossein Zare, Amirhessam Zare, Herlock Rahimi +2Structural Causal Models Counterfactual Prediction
School of Medicine, Tehran University of Medical Sciences, Tehran, Iran. · Department of Electrical and Computer Engineering, Yale University, New Haven, CT, USA. · School of Medicine, Shiraz University of Medical Sciences, Shiraz, Iran. +2
SurvPFN: Towards Foundation Models for Survival Predictions
Jun 3, 2026Samuel Böhm, Lennart Purucker, Frank Hutter +1Tabular Foundation Models Survival Analysis
Institute of Epidemiology and Prevention, Medical Center - University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany · Institute of Epidemiology and Prevention, Medical Center – University of Freiburg, Freiburg, Germany · Department of Computer Science, University of Freiburg, Freiburg, Germany +7
On the Uncertainty Quantification Ability of Tabular Foundation Models
May 31, 2026Tyler R. Johnson, Kian Ben-Jacob, Nima Negarandeh +2Tabular Foundation Models Uncertainty Quantification
Department of Mechanical and Aerospace Engineering, University of California, Irvine · Department of Civil and Environmental Engineering, University of California, Irvine
Towards Continuous-time Causal Foundation Models
May 26, 2026Dennis Thumm, Ruben Wiedemann, Ying ChenIrregular Time-Series Modeling Causal Foundation Models
National University of Singapore · Department of Mathematics, Centre for Quantitative Finance, Risk Management Institute, National University of Singapore. · Imperial College London

Oct 27, 2026J/K move · Enter open · S save

GAS-PFN: Graph-Aware Spatiotemporal Prior-Data Fitted Networks for Zero-Shot Traffic Forecasting

Joseph Natter, Long Vu, Yifan Zhang, Bobin Deng+2

Kennesaw State University · Missouri State University

Traffic Forecasting Spatiotemporal Forecasting Zero-Shot Time Series Forecasting Prior-Data Fitted Networks Spatiotemporal Transformers

Open paper AskPDF arXiv

Traffic forecasting requires modeling complex temporal dynamics and the spatial dependencies imposed by road topology. Supervised graph-based models achieve strong performance, but are often trained for a particular dataset and sensor graph, making deployment to new or evolving networks costly. Prior-Data Fitted Networks (PFNs) offer a zero-shot alternative by learning in-context inference from synthetic task priors, which provide an explicit mechanism for encoding domain-specific assumptions. Existing forecasting PFNs, however, do not explicitly represent the graph-structured dependencies governing traffic dynamics. To combine graph-aware traffic modeling with zero-shot adaptation, we introduce the Graph-Aware Spatiotemporal Prior-Data Fitted Network (GAS-PFN). GAS-PFN incorporates sampled road topology into its synthetic task prior through random-walk diffusion, teaching the model to learn graph-coupled dynamics during pretraining. Its co-designed inference architecture uses two mask-conditioned passes through a shared Transformer backbone to separately model sensor-specific temporal dynamics and spatiotemporal interactions among neighboring sensors before integrating their representations for prediction. This design enables GAS-PFN to use target-network structure and context without parameter updates. Across six benchmarks at the 90-minute horizon, GAS-PFN reduces MSE by 34.0% on average relative to TimePFN, its closest PFN-based predecessor. It also reduces MSE by 27.7% on average relative to four recent industry-developed general-purpose zero-shot forecasters. In this comparison, all models receive 90 minutes of contiguous recent history, and any additional context is drawn from each dataset's chronological training split. These results indicate that aligning PFN priors and inference architectures with graph structure improves in-context learning for zero-shot traffic forecasting.