Interventional Time Series Priors for Causal Foundation Models

• URL: http://arxiv.org/abs/2603.11090v1
• Date: Wed, 11 Mar 2026 06:14:37 GMT
• Title: Interventional Time Series Priors for Causal Foundation Models
• Abstract summary: We propose textbfCausalTimePrior, a principled framework for generating synthetic temporal structural causal models.<n>Our prior supports causal graph structures, nonlinear autoregressive mechanisms, regime-switching dynamics, and multiple intervention types.<n>We demonstrate that PFNs trained on CausalTimePrior can perform in-context causal effect estimation on held-out TSCMs.
• Score: 3.654594055714305
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Prior-data fitted networks (PFNs) have emerged as powerful foundation models for tabular causal inference, yet their extension to time series remains limited by the absence of synthetic data generators that provide interventional targets. Existing time series benchmarks generate observational data with ground-truth causal graphs but lack the interventional data required for training causal foundation models. To address this, we propose \textbf{CausalTimePrior}, a principled framework for generating synthetic temporal structural causal models (TSCMs) with paired observational and interventional time series. Our prior supports configurable causal graph structures, nonlinear autoregressive mechanisms, regime-switching dynamics, and multiple intervention types (hard, soft, time-varying). We demonstrate that PFNs trained on CausalTimePrior can perform in-context causal effect estimation on held-out TSCMs, establishing a pathway toward foundation models for time series causal inference.

Related papers

• Trio: Learning Time-Series Forecasting with Temporal-Spatial-Sample Attention and Structural Causal Priors [27.20677965052]
  Recent Prior-Data Fitted Networks (PFNs) suggest that synthetic tasks can be useful for learning transferable inference behavior.<n>We propose Trio, a sample-aware time-series forecasting architecture based on Temporal-Spatial-Sample attention.<n>We introduce a Time-Series Structural Causal Model (TS-SCM) generator to create structured synthetic forecasting tasks with dynamic lags, cross-variable interactions, noise, feedback, and distributional drift.
  arXiv  Detail & Related papers  (2026-06-05T14:01:58Z)
• TS-ICL: A Flexible Time-Indexed Foundation Model for Time Series via In-Context Learning [0.764671395172401]
  Current approaches primarily focus on forecasting, while real-world time series are often irregularly and partially observed.<n>We introduce TS-ICL, a novel probabilistic In-Context Learning encoder--regressor Transformer that unifies forecasting and imputation.
  arXiv  Detail & Related papers  (2026-06-04T08:52:21Z)
• TabCausal: Pretraining Across Causal Environments for Tabular Causal Discovery [47.81267771630766]
  Causal discovery foundation models (CDFMs) seek to amortize this problem by mapping a dataset directly to a causal graph in a single forward pass.<n>We propose TabCausal, a data-driven CDFM trained with broad causal pretraining over diverse graph priors.
  arXiv  Detail & Related papers  (2026-05-29T11:05:43Z)
• TimeAPN: Adaptive Amplitude-Phase Non-Stationarity Normalization for Time Series Forecasting [67.91113180885601]
  TimeAPN explicitly models and predicts non-stationary factors from both the time and frequency domains.<n>TimeAPN consistently improves long-term forecasting accuracy across multiple prediction horizons.<n>It outperforms state-of-the-art reversible normalization methods.
  arXiv  Detail & Related papers  (2026-03-18T07:21:24Z)
• Generating Causal Temporal Interaction Graphs for Counterfactual Validation of Temporal Link Prediction [7.925229590936017]
  We generate causal temporal interaction graphs (CTIGs) with known ground-truth causal structure.<n>To compare causal models, we propose a distance metric based on cross-model predictive error.<n>Our framework provides a foundation for causality benchmarking.
  arXiv  Detail & Related papers  (2026-02-02T14:36:18Z)
• Time Series Foundation Models for Process Model Forecasting [8.339024524110828]
  Process Model Forecasting aims to predict how the control-flow structure of a process evolves over time.<n>Machine learning and deep learning models provide only modest gains over statistical baselines.<n>We investigate Time Series Foundation Models (TSFMs) as an alternative for PMF.
  arXiv  Detail & Related papers  (2025-12-08T15:08:50Z)
• A Unified Frequency Domain Decomposition Framework for Interpretable and Robust Time Series Forecasting [81.73338008264115]
  Current approaches for time series forecasting, whether in the time or frequency domain, predominantly use deep learning models based on linear layers or transformers.<n>We propose FIRE, a unified frequency domain decomposition framework that provides a mathematical abstraction for diverse types of time series.<n>Fire consistently outperforms state-of-the-art models on long-term forecasting benchmarks.
  arXiv  Detail & Related papers  (2025-10-11T09:59:25Z)
• ARIES: Relation Assessment and Model Recommendation for Deep Time Series Forecasting [54.57031153712623]
  ARIES is a framework for assessing relation between time series properties and modeling strategies.<n>We propose the first deep forecasting model recommender, capable of providing interpretable suggestions for real-world time series.
  arXiv  Detail & Related papers  (2025-09-07T13:57:14Z)
• Learning Time-Aware Causal Representation for Model Generalization in Evolving Domains [50.66049136093248]
  We develop a time-aware structural causal model (SCM) that incorporates dynamic causal factors and the causal mechanism drifts.<n>We show that our method can yield the optimal causal predictor for each time domain.<n>Results on both synthetic and real-world datasets exhibit that SYNC can achieve superior temporal generalization performance.
  arXiv  Detail & Related papers  (2025-06-21T14:05:37Z)
• Causal Temporal Regime Structure Learning [49.77103348208835]
  We present CASTOR, a novel method that concurrently learns the Directed Acyclic Graph (DAG) for each regime.<n>We establish the identifiability of the regimes and DAGs within our framework.<n>Experiments show that CASTOR consistently outperforms existing causal discovery models.
  arXiv  Detail & Related papers  (2023-11-02T17:26:49Z)
• Add and Thin: Diffusion for Temporal Point Processes [24.4686728569167]
  ADD-THIN is a principled probabilistic denoising diffusion model for temporal point process (TPP) networks. It operates on entire event sequences and matches state-of-the-art TPP models in density estimation. In experiments on synthetic and real-world datasets, our model matches the state-of-the-art TPP models in density estimation and strongly outperforms them in forecasting.
  arXiv  Detail & Related papers  (2023-11-02T10:42:35Z)
• Supporting Optimal Phase Space Reconstructions Using Neural Network Architecture for Time Series Modeling [68.8204255655161]
  We propose an artificial neural network with a mechanism to implicitly learn the phase spaces properties. Our approach is either as competitive as or better than most state-of-the-art strategies.
  arXiv  Detail & Related papers  (2020-06-19T21:04:47Z)
• Inferring Individual Level Causal Models from Graph-based Relational Time Series [3.332377849866735]
  We formalize the problem of causal inference over graph-based relational time-series data. We propose causal inference models that leverage both the graph topology and time-series to accurately estimate local causal effects of nodes.
  arXiv  Detail & Related papers  (2020-01-16T18:48:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.