Language models infer cause and effect between event sequences
Inferring Causal Relations between Two Sequences of Events with Language Models
Artificial Intelligence
Summary
Understanding cause and effect from just a few observations is very hard for standard methods. The authors show that large language models can predict causal relations between two short sequences of events better than usual algorithms. They tested this approach on both made-up data and real-world time series. This might help detect rare problems quickly when only limited data is available.
What this means in practice
- •For alarm system operators: Identify abnormal events rapidly by inferring causal links between alarm sequences with limited observations.
- •For financial fraud detection teams: Detect suspicious transaction sequences by determining potential cause-effect relations from small data samples.
Authors
Nishchal Prasad, Eric Gaussier, Emilie Devijver, Alexander Obeid Guzman, Armen Aghasaryan, Gregor Gössler
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.