LLM-based Online Prediction of Time-varying Graph Signals

TL;DR

This paper introduces a novel LLM-based framework for predicting missing values in time-varying graph signals, demonstrating superior accuracy over traditional methods in wind-speed prediction tasks.

Contribution

It presents a new approach that uses large language models for online prediction of graph signals, integrating message-passing with LLMs to handle missing data effectively.

Findings

Outperforms online graph filtering algorithms in accuracy

Effective in predicting wind-speed graph signals

Leverages LLMs for spatial and temporal smoothness

Abstract

In this paper, we propose a novel framework that leverages large language models (LLMs) for predicting missing values in time-varying graph signals by exploiting spatial and temporal smoothness. We leverage the power of LLM to achieve a message-passing scheme. For each missing node, its neighbors and previous estimates are fed into and processed by LLM to infer the missing observations. Tested on the task of the online prediction of wind-speed graph signals, our model outperforms online graph filtering algorithms in terms of accuracy, demonstrating the potential of LLMs in effectively addressing partially observed signals in graphs.