[2412.14019] Retrieving Classes of Causal Orders with Inconsistent Knowledge Bases

Computer Science > Artificial Intelligence arXiv:2412.14019 (cs) [Submitted on 18 Dec 2024 (v1), last revised 30 Mar 2026 (this version, v4)] Title:Retrieving Classes of Causal Orders with Inconsistent Knowledge Bases Authors:Federico Baldo, Simon Ferreira, Charles K. Assaad View a PDF of the paper titled Retrieving Classes of Causal Orders with Inconsistent Knowledge Bases, by Federico Baldo and 2 other authors View PDF HTML (experimental) Abstract:Traditional causal discovery methods often depend on strong, untestable assumptions, making them unreliable in real-world applications. In this context, Large Language Models (LLMs) have emerged as a promising alternative for extracting causal knowledge from text-based metadata, effectively consolidating domain expertise. However, LLMs are prone to hallucinations, necessitating strategies that account for these limitations. One effective approach is to use a consistency measure as a proxy of reliability. Moreover, LLMs do not clearly distinguish direct from indirect causal relationships, complicating the discovery of causal Directed Acyclic Graphs (DAGs), which are often sparse. This ambiguity is evident in the way informal sentences are formulated in various domains. For this reason, focusing on causal orders provides a more practical and direct task for LLMs. We propose a new method for deriving abstractions of causal orders that maximizes a consistency score obtained from an LLM. Our approach begins by computing pairwise con...