# Learning to Defer for Causal Discovery with Imperfect Experts

arXiv: https://arxiv.org/abs/2502.13132 · HTML: https://mlsanity.com/paper/2502.13132/

## 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 (most cited first, 19 of 19)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Causal Reasoning and Large Language Models: Opening a New Frontier for Causality](https://mlsanity.com/paper/2305.00050/) (2023)
- [Consistent Estimators for Learning to Defer to an Expert](https://mlsanity.com/paper/2006.01862/) (2020)
- [Causal Discovery with Continuous Additive Noise Models](https://mlsanity.com/paper/1309.6779/) (2013)
- [Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer](https://mlsanity.com/paper/1711.06664/) (2017)
- [CAM: Causal additive models, high-dimensional order search and penalized regression](https://mlsanity.com/paper/1310.1533/) (2013)
- [Causal Inference and Causal Explanation with Background Knowledge](https://mlsanity.com/paper/1302.4972/) (2013)
- [Causal Parrots: Large Language Models May Talk Causality But Are Not Causal](https://mlsanity.com/paper/2308.13067/) (2023)
- [Can Large Language Models Infer Causation from Correlation?](https://mlsanity.com/paper/2306.05836/) (2023)
- [LMPriors: Pre-Trained Language Models as Task-Specific Priors](https://mlsanity.com/paper/2210.12530/) (2022)
- [Integrating Large Language Model for Improved Causal Discovery](https://mlsanity.com/paper/2306.16902/) (2023)
- [The impact of prior knowledge on causal structure learning](https://mlsanity.com/paper/2102.00473/) (2021)
- [ALCM: Autonomous LLM-Augmented Causal Discovery Framework](https://mlsanity.com/paper/2405.01744/) (2024)
- [Ancestral Causal Inference](https://mlsanity.com/paper/1606.07035/) (2016)
- [Minimax Rates and Efficient Algorithms for Noisy Sorting](https://mlsanity.com/paper/1710.10388/) (2017)
- [Large Language Models for Constrained-Based Causal Discovery](https://mlsanity.com/paper/2406.07378/) (2024)
- [On Sample Complexity Upper and Lower Bounds for Exact Ranking from Noisy Comparisons](https://mlsanity.com/paper/1909.03194/) (2019)
- [Expert-Aided Causal Discovery of Ancestral Graphs](https://mlsanity.com/paper/2309.12032/) (2023)
- [An Explainable AI Approach to Large Language Model Assisted Causal Model Auditing and Development](https://mlsanity.com/paper/2312.16211/) (2023)

## Built on by (most cited first, 2 of 2)

- [Realizing LLMs' Causal Potential Requires Science-Grounded, Novel Benchmarks](https://mlsanity.com/paper/2510.16530/) (2025)
- [xWhyL: Causal Interactive Learning](https://mlsanity.com/paper/2609.26037/) (2026)

## 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.