[2603.04276] Causality Elicitation from Large Language Models

Computer Science > Machine Learning arXiv:2603.04276 (cs) [Submitted on 4 Mar 2026] Title:Causality Elicitation from Large Language Models Authors:Takashi Kameyama, Masahiro Kato, Yasuko Hio, Yasushi Takano, Naoto Minakawa View a PDF of the paper titled Causality Elicitation from Large Language Models, by Takashi Kameyama and Masahiro Kato and Yasuko Hio and Yasushi Takano and Naoto Minakawa View PDF HTML (experimental) Abstract:Large language models (LLMs) are trained on enormous amounts of data and encode knowledge in their parameters. We propose a pipeline to elicit causal relationships from LLMs. Specifically, (i) we sample many documents from LLMs on a given topic, (ii) we extract an event list from from each document, (iii) we group events that appear across documents into canonical events, (iv) we construct a binary indicator vector for each document over canonical events, and (v) we estimate candidate causal graphs using causal discovery methods. Our approach does not guarantee real-world causality. Rather, it provides a framework for presenting the set of causal hypotheses that LLMs can plausibly assume, as an inspectable set of variables and candidate graphs. Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Econometrics (econ.EM) Cite as: arXiv:2603.04276 [cs.LG]   (or arXiv:2603.04276v1 [cs.LG] for this version)   https://doi.org/10.48550/arXiv.2603.04276 Focus to learn more arXiv-issued DOI via DataCite (pen...