mlsanityData to 14 Oct 2026

Learning to Defer for Causal Discovery with Imperfect Experts

arXiv 2502.13132 ↗First posted 18 Feb 2025cs.LG cs.AI stat.MLcited by 2

Not on the map. Why?

Abstract (arXiv)

Integrating expert knowledge, e.g. from large language models, into causal discovery algorithms can be challenging when the knowledge is not guaranteed to be correct. Expert recommendations may contradict data-driven results, and their reliability can vary significantly depending on the domain or specific query. Existing methods based on soft constraints or inconsistencies in predicted causal relationships fail to account for these variations in expertise. To remedy this, we propose L2D-CD, a method for gauging the correctness of expert recommendations and optimally combining them with data-driven causal discovery results. By adapting learning-to-defer (L2D) algorithms for pairwise causal discovery (CD), we learn a deferral function that selects whether to rely on classical causal discovery methods using numerical data or expert recommendations based on textual meta-data. We evaluate L2D-CD on the canonical T"ubingen pairs dataset and demonstrate its superior performance compared to both the causal discovery method and the expert used in isolation. Moreover, our approach identifies domains where the expert's performance is strong or weak. Finally, we outline a strategy for generalizing this approach to causal discovery on graphs with more than two variables, paving the way for further research in this area.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 19 of its 54 references that are in the map. First posted on arXiv 18 Feb 2025.

Builds on

19 of 54 references.

References · 19

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2023Causal Reasoning and Large Language Models: Opening a New Frontier for Causalitycited by 353
  3. 2020Consistent Estimators for Learning to Defer to an Expertcited by 271
  4. 2013Causal Discovery with Continuous Additive Noise Modelscited by 255
  5. 2017Predict Responsibly: Improving Fairness and Accuracy by Learning to Defercited by 233
  6. 2013CAM: Causal additive models, high-dimensional order search and penalized regressioncited by 178
  7. 2013Causal Inference and Causal Explanation with Background Knowledgecited by 170
  8. 2023Causal Parrots: Large Language Models May Talk Causality But Are Not Causalcited by 150
  9. 2023Can Large Language Models Infer Causation from Correlation?cited by 146
  10. 2022LMPriors: Pre-Trained Language Models as Task-Specific Priorscited by 56
  11. 2023Integrating Large Language Model for Improved Causal Discoverycited by 54
  12. 2021The impact of prior knowledge on causal structure learningcited by 35
7 more
  1. 2024ALCM: Autonomous LLM-Augmented Causal Discovery Frameworkcited by 29
  2. 2016Ancestral Causal Inferencecited by 24
  3. 2017Minimax Rates and Efficient Algorithms for Noisy Sortingcited by 15
  4. 2024Large Language Models for Constrained-Based Causal Discoverycited by 15
  5. 2019On Sample Complexity Upper and Lower Bounds for Exact Ranking from Noisy Comparisonscited by 10
  6. 2023Expert-Aided Causal Discovery of Ancestral Graphscited by 8
  7. 2023An Explainable AI Approach to Large Language Model Assisted Causal Model Auditing and Developmentcited by 5

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 1.

2 papers.

  1. 2025Realizing LLMs' Causal Potential Requires Science-Grounded, Novel Benchmarkscited by 4
  2. 2026xWhyL: Causal Interactive Learningcited by 0