mlsanityData to 7 Oct 2026

Test-Time Learning of Causal Structure from Interventional Data

arXiv 2602.19131 ↗First posted 22 Feb 2026cs.LG cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Supervised causal learning has shown promise in causal discovery, yet it often struggles with generalization across diverse interventional settings, particularly when intervention targets are unknown. To address this, we propose TICL (Test-time Interventional Causal Learning), a novel method that synergizes Test-Time Training with Joint Causal Inference. Specifically, we design a self-augmentation strategy to generate instance-specific training data at test time, effectively avoiding distribution shifts. Furthermore, by integrating joint causal inference, we developed a PC-inspired two-phase supervised learning scheme, which effectively leverages self-augmented training data while ensuring theoretical identifiability. Extensive experiments on bnlearn benchmarks demonstrate TICL's superiority in multiple aspects of causal discovery and intervention target detection.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 43 of its 93 references that are in the map. First posted on arXiv 22 Feb 2026.

Builds on

43 of 93 references.

References · 43

  1. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  2. 2018DAGs with NO TEARS: Continuous Optimization for Structure Learningcited by 745
  3. 2022Continual Test-Time Domain Adaptationcited by 520
  4. 2019DAG-GNN: DAG Structure Learning with Graph Neural Networkscited by 329
  5. 2024Learning to (Learn at Test Time): RNNs with Expressive Hidden Statescited by 329
  6. 2020Differentiable Causal Discovery from Interventional Datacited by 218
  7. 2020On the Role of Sparsity and DAG Constraints for Learning Linear DAGscited by 186
  8. 2013Causal Inference and Causal Explanation with Background Knowledgecited by 168
  9. 2022Bayesian Structure Learning with Generative Flow Networkscited by 153
  10. 2019Learning Neural Causal Models from Unknown Interventionscited by 113
  11. 2023Nonparametric Identifiability of Causal Representations from Unknown Interventionscited by 105
  12. 2025One-Minute Video Generation with Test-Time Trainingcited by 100
31 more
  1. 2022Amortized Inference for Causal Structure Learningcited by 89
  2. 2023Identifiability Guarantees for Causal Disentanglement from Soft Interventionscited by 84
  3. 2021Efficient Neural Causal Discovery without Acyclicity Constraintscited by 82
  4. 2023Test-Time Training on Nearest Neighbors for Large Language Modelscited by 82
  5. 2017Permutation-based Causal Inference Algorithms with Interventionscited by 78
  6. 2018Characterizing and Learning Equivalence Classes of Causal DAGs under Interventionscited by 76
  7. 2013Learning Equivalence Classes of Bayesian Networks Structurescited by 75
  8. 2019Permutation-Based Causal Structure Learning with Unknown Intervention Targetscited by 69
  9. 2022Learning to Induce Causal Structurecited by 68
  10. 2023A Survey on Causal Discovery: Theory and Practicecited by 63
  11. 2019A Graph Autoencoder Approach to Causal Structure Learningcited by 55
  12. 2021Ordering-Based Causal Discovery with Reinforcement Learningcited by 55
  13. 2025End-to-End Test-Time Training for Long Contextcited by 53
  14. 2013Causal Discovery from Changescited by 52
  15. 2015Partition MCMC for inference on acyclic digraphscited by 33
  16. 2023BayesDAG: Gradient-Based Posterior Inference for Causal Discoverycited by 33
  17. 2022BaCaDI: Bayesian Causal Discovery with Unknown Interventionscited by 26
  18. 2020Supervised Whole DAG Causal Discoverycited by 19
  19. 2024Sample, estimate, aggregate: A recipe for causal discovery foundation modelscited by 19
  20. 2020Meta Learning for Causal Directioncited by 16
  21. 2021ML4C: Seeing Causality Through Latent Vicinitycited by 11
  22. 2022GFlowCausal: Generative Flow Networks for Causal Discoverycited by 11
  23. 2014The Randomized Causation Coefficientcited by 10
  24. 2021Scalable Intervention Target Estimation in Linear Modelscited by 10
  25. 2022Causal discovery for observational sciences using supervised machine learningcited by 10
  26. 2025Learning with Calibration: Exploring Test-Time Computing of Spatio-Temporal Forecastingcited by 9
  27. 2016A Review on Algorithms for Constraint-based Causal Discoverycited by 8
  28. 2023Learning Unknown Intervention Targets in Structural Causal Models from Heterogeneous Datacited by 7
  29. 2023Conditional Independence Testing with Heteroskedastic Data and Applications to Causal Discoverycited by 6
  30. 2024CORE: Towards Scalable and Efficient Causal Discovery with Reinforcement Learningcited by 6
  31. 2023Bayesian causal discovery from unknown general interventionscited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Test Time Training for Supervised Causal Learningcited by 1
  2. 2026Demistifying Data and Simulator Assumptions in Supervised Causal Discoverycited by 0