Large Language Models for Causal Discovery: Current Landscape and Future Directions

2024-02-16

AI Summary

This work addresses the integration of large language models (LLMs) into causal discovery (CD). We systematically investigate three synergistic pathways: (1) direct extraction of causal relations from unstructured text; (2) injection of domain knowledge into statistical causal inference methods to enhance interpretability and reliability; and (3) optimization of causal graph structure learning. Methodologically, we introduce the first unified analytical framework for LLM-driven CD, proposing a novel metadata- and natural-language-coordinated causal reasoning paradigm. We further establish the first dedicated evaluation benchmark and testing protocol for LLM-based CD. Our empirical analysis characterizes LLMs as “imperfect causal experts,” rigorously delineating their capabilities and limitations while identifying critical research gaps. The results provide both a methodological foundation and practical guidance for developing next-generation causal AI systems that are knowledge-augmented, robust, and inherently interpretable.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Natural Language Processing: (Large) Language Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Causal discovery (CD) and Large Language Models (LLMs) have emerged as transformative fields in artificial intelligence that have evolved largely independently. While CD specializes in uncovering cause-effect relationships from data, and LLMs excel at natural language processing and generation, their integration presents unique opportunities for advancing causal understanding. This survey examines how LLMs are transforming CD across three key dimensions: direct causal extraction from text, integration of domain knowledge into statistical methods, and refinement of causal structures. We systematically analyze approaches that leverage LLMs for CD tasks, highlighting their innovative use of metadata and natural language for causal inference. Our analysis reveals both LLMs' potential to enhance traditional CD methods and their current limitations as imperfect expert systems. We identify key research gaps, outline evaluation frameworks and benchmarks for LLM-based causal discovery, and advocate future research efforts for leveraging LLMs in causality research. As the first comprehensive examination of the synergy between LLMs and CD, this work lays the groundwork for future advances in the field.

Problem

• Integrating LLMs with causal discovery
• Enhancing causal inference using metadata
• Identifying research gaps in LLM-based CD

Innovation

• LLMs enhance causal discovery
• Integrate domain knowledge with statistics
• Refine causal structures using metadata