mlsanityData to 7 Oct 2026

CausalSteward: An Agentic Divide-Conquer-Combine Copilot for Causal Discovery

arXiv 2607.01936 ↗First posted 2 Jul 2026cs.MA cs.AIcited by 0

On the map: LLM agents and tool use → Medical and scientific domain agents

Abstract (arXiv)

Learning causal models from high-dimensional data is a significant challenge, particularly in real-world settings where violations of core assumptions lead to causal identifiability issues. Although massive amounts of prior knowledge are available, and contain valuable causal information, effectively integrating this knowledge into the causal discovery process remains an open problem. We introduce CausalSTeward (CAST), a novel human-in-the-loop framework for interactively assembling large causal models. CausalSteward is a multi-agent collaborative system that tackles high-dimensional causality through a divide-and-conquer approach where large clusters of variables are iteratively partitioned and then separately analyzed. Our framework fuses prior knowledge with a data-driven approach by using tailored tools such as retrieval augmented generation and conditional independence tests. Finally, we use this work to examine the capabilities and limitations of causal reasoning in multi-agent frameworks, and how the human-in-the-loop can contribute to accurate and trustworthy results.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Medical and scientific domain agents, one of 16 directions in LLM agents and tool use. Builds on 30 of its 57 references that are in the map. First posted on arXiv 2 Jul 2026.

Builds on

30 of 57 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Causal-Copilot: An Autonomous Causal Analysis Agent17 Apr
  2. 2024Multi-Agent Causal Discovery Using Large Language Models21 Jul

Other references · 28

  1. 2024GPT-4o System Cardcited by 4,467
  2. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  3. 2020Differentiable Causal Discovery from Interventional Datacited by 218
  4. 2023Talk like a Graph: Encoding Graphs for Large Language Modelscited by 175
  5. 2023Causal Parrots: Large Language Models May Talk Causality But Are Not Causalcited by 146
  6. 2023Integrating Large Language Model for Improved Causal Discoverycited by 54
  7. 2024Efficient Causal Graph Discovery Using Large Language Modelscited by 51
  8. 2023Causal-Discovery Performance of ChatGPT in the context of Neuropathic Pain Diagnosiscited by 36
  9. 2019Neuropathic Pain Diagnosis Simulator for Causal Discovery Algorithm Evaluationcited by 32
  10. 2024Integrating Large Language Models in Causal Discovery: A Statistical Causal Approachcited by 31
  11. 2022Can Foundation Models Talk Causality?cited by 30
  12. 2021Iterative Causal Discovery in the Possible Presence of Latent Confounders and Selection Biascited by 21
16 more
  1. 2024Sample, estimate, aggregate: A recipe for causal discovery foundation modelscited by 19
  2. 2024The Landscape of Causal Discovery Data: Grounding Causal Discovery in Real-World Applicationscited by 17
  3. 2018Causal Identification under Markov Equivalencecited by 15
  4. 2023Zero-shot Causal Graph Extrapolation from Text via LLMscited by 15
  5. 2013On the Equivalence of Causal Modelscited by 13
  6. 2024An Examination on the Effectiveness of Divide-and-Conquer Prompting in Large Language Modelscited by 12
  7. 2024The Causal Chambers: Real Physical Systems as a Testbed for AI Methodologycited by 12
  8. 2024Teaching Transformers Causal Reasoning through Axiomatic Trainingcited by 12
  9. 2024CausalGraph2LLM: Evaluating LLMs for Causal Queriescited by 12
  10. 2023Expert-Aided Causal Discovery of Ancestral Graphscited by 8
  11. 2024DCILP: A Distributed Approach for Large-Scale Causal Structure Learningcited by 8
  12. 2023Mitigating Prior Errors in Causal Structure Learning: A Resilient Approach via Bayesian Networkscited by 5
  13. 2023Causal discovery using dynamically requested knowledgecited by 5
  14. 2024Local Causal Discovery with Background Knowledgecited by 5
  15. 2025CausalMan: A physics-based simulator for large-scale causalitycited by 4
  16. 2023On the Unlikelihood of D-Separationcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2023Towards CausalGPT: A Multi-Agent Approach for Faithful Knowledge Reasoning via Promoting Causal Consistency in LLMs
  2. 2025Causal Reasoning in Pieces: Modular In-Context Learning for Causal Discovery
  3. 2025Causal MAS: A Survey of Large Language Model Architectures for Discovery and Effect Estimation
  4. 2026Step-by-Step Causality: Transparent Causal Discovery with Multi-Agent Tree-Query and Adversarial Confidence Estimation
  5. 2026CausalArena: Benchmarking Causal Discovery in the Foundation Model Era
  6. 2024Large Language Models for Constrained-Based Causal Discovery
  7. 2026Traceable Latent Variable Discovery Based on Multi-Agent Collaboration
  8. 2026Causal Discovery in the Era of Agents