CausalArena: Benchmarking Causal Discovery in the Foundation Model Era

• URL: http://arxiv.org/abs/2609.11897v1
• Date: Thu, 10 Sep 2026 17:53:15 GMT
• Title: CausalArena: Benchmarking Causal Discovery in the Foundation Model Era
• Abstract summary: CausalArena is a unified benchmark for causal discovery under a common protocol.<n>We show that strong performance in one benchmark regime does not reliably transfer to others.
• Score: 47.81267771630766
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Causal discovery aims to uncover causal structures from data and is fundamental to scientific reasoning and intervention-based decision making. Its evaluation relies heavily on structural causal models (SCMs), which specify a causal graph together with the mechanisms that generate data, yet existing studies differ substantially in graph families, mechanisms, and evaluation protocols. The emergence of causal discovery foundation models (CDFMs) further complicates evaluation: performance may reflect not only causal discovery ability, but also overlap between pretraining environments and test SCMs, making results on fixed synthetic benchmarks difficult to interpret. We introduce CausalArena, a unified and evolvable benchmark for causal discovery under a common protocol. Synthetic SCMs supply controlled breadth over structures and mechanisms; semantic operational SCMs provide human-auditable, semantically grounded environments beyond standard synthetic generators; and formula-grounded SCMs test discovery under explicit scientific mechanisms. Public real-world datasets provide an additional external-validity check. Experiments across classical, neural, and pretrained methods reveal substantial ranking shifts across SCM families and protocols, showing that strong performance in one benchmark regime does not reliably transfer to others. These results highlight benchmark diversity and pretraining--evaluation overlap as central challenges for evaluating causal discovery in the foundation model era.

Related papers

• DAG-FM: A Foundation Model for Causal Discovery under Heterogeneous Causal Mechanisms [48.57846236686172]
  We propose textbfDAG-FM, a novel foundation model architecture that amortizes causal discovery.<n>Unlike direct matrix prediction, DAG-FM decomposes the causal discovery process into two auto-regressive stages.<n>We show that DAG-FM achieves state-of-the-art performance on both synthetic benchmarks and complex real-world datasets.
  arXiv  Detail & Related papers  (2026-07-13T13:00:06Z)
• CDFM: Towards a General-Purpose Causal Discovery Foundation Model [62.32138962508216]
  Causal discovery is the process of recovering underlying causal structures from observational data.<n>We formulate the Causal Discovery Foundation Model (CDFM) as a unified, general-purpose framework for zero-shot structural inference.<n>CDFM consistently outperforms traditional algorithms, driving a paradigm shift toward a general-purpose causal discovery foundation model.
  arXiv  Detail & Related papers  (2026-07-13T12:58:28Z)
• Causal Ensemble Agent: Hierarchical Causal Discovery with LLM-guided Expert Reweighting [95.24017293537025]
  Causal Ensemble Agent (CEA) is a novel framework that aggregates structural insights from statistical discovery experts across different graph levels.<n>CEA achieves the strongest overall performance across a wide range of causal discovery methods.
  arXiv  Detail & Related papers  (2026-06-09T09:09:07Z)
• 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)
• PRIM: Meta-Learned Bayesian Root Cause Analysis [6.537088094470879]
  We introduce PRIM (Prior-fitted Root cause Identification with Meta-learning), a causal meta-learning approach.<n> PRIM implicitly identifies changes in the data-generating mechanism between baseline and anomalous periods.<n>It infers distributional differences without explicit statistical testing, and implicitly learns causal structure without model fitting at test time.
  arXiv  Detail & Related papers  (2026-05-09T08:14:09Z)
• Kolmogorov-Arnold causal generative models [40.44169036529174]
  Causal generative models provide a principled framework for answering observational, interventional, and counterfactual queries from observational data.<n>We propose KaCGM, a causal generative model for mixed-type tabular data where each structural equation is parameterized by a Kolmogorov--Arnold Network (KAN)<n>This decomposition enables direct inspection of learned causal mechanisms, including symbolic approximations and visualization of parent--child relationships.
  arXiv  Detail & Related papers  (2026-03-20T17:58:38Z)
• dcFCI: Robust Causal Discovery Under Latent Confounding, Unfaithfulness, and Mixed Data [1.9797215742507548]
  We introduce the first nonparametric score to assess a Partial Ancestral Graph's compatibility with observed data.<n>We then propose data-compatible Fast Causal Inference (dcFCI) to jointly address latent confounding, empirical unfaithfulness, and mixed data types.
  arXiv  Detail & Related papers  (2025-05-10T07:05:19Z)
• CausalBench: A Large-scale Benchmark for Network Inference from Single-cell Perturbation Data [61.088705993848606]
  We introduce CausalBench, a benchmark suite for evaluating causal inference methods on real-world interventional data. CaulBench incorporates biologically-motivated performance metrics, including new distribution-based interventional metrics.
  arXiv  Detail & Related papers  (2022-10-31T13:04:07Z)

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.