---
title: "Towards Comprehensive Benchmarking of Domain Generalization and Adaptation for Time-Series Forecasting in Clinical Multivariate Time-Series Forecasting"
authors:
  - "Mayra Elwes"
doi: "10.24963/ijcai.2026/942"
url: "https://ijcai.org/proceedings/2026/942"
pdf: "https://ijcai.org/proceedings/2026/0942.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "8327-8328"
track: "Doctoral Consortium"
keywords:
  - "Multidisciplinary Topics and Applications: Health and medicine"
  - "Machine Learning: Evaluation"
  - "AI Ethics, Trust, Fairnes: Safety and robustness"
  - "AI: Machine Learning"
---

# Towards Comprehensive Benchmarking of Domain Generalization and Adaptation for Time-Series Forecasting in Clinical Multivariate Time-Series Forecasting

Mayra Elwes

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 8327-8328. DOI: [10.24963/ijcai.2026/942](https://doi.org/10.24963/ijcai.2026/942)

## Abstract

Deep learning models for clinical time-series forecasting face significant challenges in generalizability when deployed across different clinical settings. Various choices in the deep learning pipeline can affect a model's out-of-distribution performance. This abstract introduces a comprehensive benchmarking framework to assess domain generalization and adaptation under varying distribution shifts in multivariate time-series forecasting using real-world data. Preliminary empirical results on the influence of backbone model architecture and a proxy metric, the robustness slope, are presented alongside remaining challenges.