mlsanityData to 7 Oct 2026

DMCD: Semantic-Statistical Framework for Causal Discovery

arXiv 2602.20333 ↗First posted 23 Feb 2026cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

We present DMCD (DataMap Causal Discovery), a two-phase causal discovery framework that integrates LLM-based semantic drafting from variable metadata with statistical validation on observational data. In Phase I, a large language model proposes a sparse draft DAG, serving as a semantically informed prior over the space of possible causal structures. In Phase II, this draft is audited and refined via conditional independence testing, with detected discrepancies guiding targeted edge revisions. We evaluate our approach on three metadata-rich real-world benchmarks spanning industrial engineering, environmental monitoring, and IT systems analysis. Across these datasets, DMCD achieves competitive or leading performance against diverse causal discovery baselines, with particularly large gains in recall and F1 score. Probing and ablation experiments suggest that these improvements arise from semantic reasoning over metadata rather than memorization of benchmark graphs. Overall, our results demonstrate that combining semantic priors with principled statistical verification yields a high-performing and practically effective approach to causal structure learning.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 27 of its 36 references that are in the map. First posted on arXiv 23 Feb 2026.

Builds on

27 of 36 references.

References · 27

  1. 2023Causal Reasoning and Large Language Models: Opening a New Frontier for Causalitycited by 348
  2. 2022CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learningcited by 342
  3. 2023RLTF: Reinforcement Learning from Unit Test Feedbackcited by 110
  4. 2022Deep End-to-end Causal Inferencecited by 93
  5. 2023Can large language models build causal graphs?cited by 72
  6. 2023Causal Discovery with Language Models as Imperfect Expertscited by 62
  7. 2023Integrating Large Language Model for Improved Causal Discoverycited by 54
  8. 2024Efficient Causal Graph Discovery Using Large Language Modelscited by 51
  9. 2023Causal Order: The Key to Leveraging Imperfect Experts in Causal Inferencecited by 46
  10. 2023CausalTime: Realistically Generated Time-series for Benchmarking of Causal Discoverycited by 35
  11. 2024Integrating Large Language Models in Causal Discovery: A Statistical Causal Approachcited by 31
  12. 2024Causal Inference with Large Language Model: A Surveycited by 31
15 more
  1. 2022Can Foundation Models Talk Causality?cited by 30
  2. 2024ALCM: Autonomous LLM-Augmented Causal Discovery Frameworkcited by 29
  3. 2024Large Language Models for Causal Discovery: Current Landscape and Future Directionscited by 25
  4. 2024Large Language Models are Effective Priors for Causal Graph Discoverycited by 23
  5. 2023pgmpy: A Python Toolkit for Bayesian Networkscited by 21
  6. 2023Causal Structure Learning Supervised by Large Language Modelcited by 21
  7. 2023Case Studies of Causal Discovery from IT Monitoring Time Seriescited by 14
  8. 2024RealTCD: Temporal Causal Discovery from Interventional Data with Large Language Modelcited by 13
  9. 2024Exploring Multi-Modal Data with Tool-Augmented LLM Agents for Precise Causal Discoverycited by 11
  10. 2022CIPCaD-Bench: Continuous Industrial Process datasets for benchmarking Causal Discovery methodscited by 10
  11. 2021Interactive Causal Structure Discovery in Earth System Sciencescited by 6
  12. 2024LLM-initialized Differentiable Causal Discoverycited by 6
  13. 2025Reinforcing Code Generation: Improving Text-to-SQL with Execution-Based Learningcited by 5
  14. 2022A Simple Unified Approach to Testing High-Dimensional Conditional Independences for Categorical and Ordinal Datacited by 4
  15. 2025Large Language Models for Zero-shot Inference of Causal Structures in Biologycited by 2

Built on by

Not cited yet.