Inferring Causal Relations between Two Sequences of Events with Language Models

2026-09-30

AI Summary

This study addresses the challenge that traditional causal discovery algorithms rely heavily on large-scale data, rendering them ineffective in extremely data-scarce scenarios involving only a single observational sequence. To overcome this limitation, this work pioneers the integration of large language models (LLMs) into single-sequence causal inference, leveraging their robust predictive and reasoning capabilities to identify latent causal structures from merely two event sequences. By circumventing the data bottlenecks inherent in conventional methods, the proposed approach establishes a new paradigm for time-series causal analysis under small-sample and information-constrained conditions. Experiments on both synthetic and real-world datasets demonstrate that our method significantly outperforms standard causal discovery baselines while effectively detecting rare anomalous events.

Abstract

Causal AI is a branch of Artificial Intelligence which helps understand and reason about cause and effect relationships, not just patterns or correlations. Causal discovery aims to infer elements of the underlying causal structure--often represented as a directed graph--from observational and, when available, interventional data. While causal discovery is the fundamental step for moving beyond mere associations toward genuine understanding, and thus the basic building block of causal AI, it becomes intrinsically difficult when causal relations must be inferred from single observations. In such situations, standard causal discovery methods cannot be used and one has to identify causal relations from limited amount of information. This is typically the case for, e.g., sequences of events produced by different alarms which need to be analyzed on the fly to detect abnormal phenomena, which are usually rare. We show in this study that it is possible to leverage the predictive power of Large Language Models (LLMs) to infer causal relations between only two sequences of events. This approach, which is validated on both synthetic and real data, provides better results than standard causal discovery algorithms on several time series data, even though these data were converted into smaller, single observed sequences.

Problem

• Causal Discovery
• Event Sequences
• Single Observations
• Causal Relations
• Anomaly Detection

Innovation

• Causal Discovery
• Large Language Models
• Event Sequences
• Causal AI
• Time Series