Deep Research for Forecasting

ServiceNow develops a platform for client organizations to manage and automate large-scale processes across various industries. In 2020, ServiceNow acquired Element AI to strengthen its presence in the Artificial Intelligence (AI) research landscape and the Canadian AI ecosystem. This acquisition enabled the developmentof AI-driven products that improve the platform’s capabilities. ServiceNow Research has made significant contributions to the field of foundation models, notably in Natural Language Processing (NLP), and has a strong presence in developing generative models for different data domains.
This project addresses the challenge of improving time-series forecasting by extending the static knowledge base of Large Language Models (LLMs). It aims to enable AI agents to reason, discover, and retrieve crucial side information, assess uncertainty, and adapt Deep Research methods for time-series analysis. The goal is to develop an Agentic Forecaster that can identify anomalies, triggers, and influential covariates, refining forecasts and enhancing interpretability and reliability.
Expanding deep research agents for time-series forecasting improves predictive accuracy, enabling better resource allocation, reduced waste, and optimized investment strategies. These advancements lower uncertainty, stimulate innovation, and create high-value analytical services supporting economic growth. To encourage this line of research, the findings of this work will be open sourced.

Faculty Supervisor:

Lijun Sun

Student:

Partner:

ServiceNow Canada

Discipline:

Computer science

Sector:

Professional, scientific and technical services

University:

McGill University

Program: