mlsanityData to 14 Oct 2026

Large Language Models are Effective Priors for Causal Graph Discovery

arXiv 2405.13551 ↗First posted 22 May 2024cs.LG cs.AIcited by 24

Not on the map. Why?

Abstract (arXiv)

Causal structure discovery from observations can be improved by integrating background knowledge provided by an expert to reduce the hypothesis space. Recently, Large Language Models (LLMs) have begun to be considered as sources of prior information given the low cost of querying them relative to a human expert. In this work, firstly, we propose a set of metrics for assessing LLM judgments for causal graph discovery independently of the downstream algorithm. Secondly, we systematically study a set of prompting designs that allows the model to specify priors about the structure of the causal graph. Finally, we present a general methodology for the integration of LLM priors in graph discovery algorithms, finding that they help improve performance on common-sense benchmarks and especially when used for assessing edge directionality. Our work highlights the potential as well as the shortcomings of the use of LLMs in this problem space.

Key facts. Cited by 24 papers in the map (data through 14 Oct 2026). Builds on 18 of its 37 references that are in the map. First posted on arXiv 22 May 2024.

Builds on

18 of 37 references.

References · 18

  1. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  2. 2018MINE: Mutual Information Neural Estimationcited by 683
  3. 2020Scaling Laws for Autoregressive Generative Modelingcited by 522
  4. 2023Causal Reasoning and Large Language Models: Opening a New Frontier for Causalitycited by 353
  5. 2023On the Planning Abilities of Large Language Models : A Critical Investigationcited by 312
  6. 2017Thinking Fast and Slow with Deep Learning and Tree Searchcited by 252
  7. 2013Causal Inference and Causal Explanation with Background Knowledgecited by 170
  8. 2023Can Large Language Models Infer Causation from Correlation?cited by 146
  9. 2023Can large language models build causal graphs?cited by 73
  10. 2023Causal Discovery with Language Models as Imperfect Expertscited by 64
  11. 2022LMPriors: Pre-Trained Language Models as Task-Specific Priorscited by 56
  12. 2023Integrating Large Language Model for Improved Causal Discoverycited by 54
6 more
  1. 2023Causal Order: The Key to Leveraging Imperfect Experts in Causal Inferencecited by 48
  2. 2022Can Foundation Models Talk Causality?cited by 30
  3. 2020Causal BERT : Language models for causality detection between events expressed in textcited by 22
  4. 2021Typing assumptions improve identification in causal discoverycited by 10
  5. 2021Planning Spatial Networks with Monte Carlo Tree Searchcited by 5
  6. 2023Tree Search in DAG Space with Model-based Reinforcement Learning for Causal Discoverycited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 8; 2026: 14.

24 papers.

Sort:

By area:

  1. 2024Multi-Agent Causal Discovery Using Large Language Modelscited by 26
  2. 2025Large language models for artificial general intelligence (AGI): A survey of foundational principles and approachescited by 18
  3. 2025A Survey on Enhancing Causal Reasoning Ability of Large Language Modelscited by 8
  4. 2025LLM Cannot Discover Causality, and Should Be Restricted to Non-Decisional Support in Causal Discoverycited by 8
  5. 2024LLM-initialized Differentiable Causal Discoverycited by 6
  6. 2026CausaLab: A Scalable Environment for Interactive Causal Discovery Toward AI Scientistscited by 4
  7. 2025Revealing Multimodal Causality with Large Language Modelscited by 3
  8. 2025Score-informed Neural Operator for Enhancing Ordering-based Causal Discoverycited by 2
  9. 2025Incorporating Expert Knowledge into Bayesian Causal Discovery of Mixtures of Directed Acyclic Graphscited by 2
  10. 2026IV Co-Scientist: Multi-Agent LLM Framework for Causal Instrumental Variable Discoverycited by 2
  11. 2026DCFA: Dual-view Causal-inspired Attribution for Failure Reasoning in LLM-based Multi-agent Systemscited by 2
  12. 2025Can LLMs Assist Expert Elicitation for Probabilistic Causal Modeling?cited by 1
12 more
  1. 2025CARE: Turning LLMs Into Causal Reasoning Expertcited by 1
  2. 2026KG-SoftMAP: Soft Knowledge-Graph Priors for Bayesian Network Structure Learning from Sparse Discrete Datacited by 1
  3. 2026Step-by-Step Causality: Transparent Causal Discovery with Multi-Agent Tree-Query and Adversarial Confidence Estimationcited by 0
  4. 2026DMCD: Semantic-Statistical Framework for Causal Discoverycited by 0
  5. 2026From Prompts to Tokens: Internalizing Causal Supervision in Vision-Language Model for Multi-Image Causal Reasoningcited by 0
  6. 2026Structured Inference with Large Language Gibbscited by 0
  7. 2026Causal Discovery in the Era of Agentscited by 0
  8. 2026DKCD: Domain Knowledge-Enhanced Causal Discovery from Unstructured Datacited by 0
  9. 2026EviDAG: Auditable Causal DAG Authoring with Biomedical Literaturecited by 0
  10. 2026Guide, Not Bind: Why Defeasible Priors Fail in Augmented Lagrangian Causal Discoverycited by 0
  11. 2026ABSOL: Aggregated Bayesian Subsampling Orchestrated with LLMscited by 0
  12. 2026Coupling Noisy Pairwise Knowledge to the DAG Posterior for Causal Discoverycited by 0