RobuMTL: Enhancing Multi-Task Learning Robustness Against Weather Conditions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shaffee, Tasneem, Reda, Sherief
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912827652964352
author Shaffee, Tasneem
Reda, Sherief
author_facet Shaffee, Tasneem
Reda, Sherief
contents Robust Multi-Task Learning (MTL) is crucial for autonomous systems operating in real-world environments, where adverse weather conditions can severely degrade model performance and reliability. In this paper, we introduce RobuMTL, a novel architecture designed to adaptively address visual degradation by dynamically selecting task-specific hierarchical Low-Rank Adaptation (LoRA) modules and a LoRA expert squad based on input perturbations in a mixture-of-experts fashion. Our framework enables adaptive specialization based on input characteristics, improving robustness across diverse real-world conditions. To validate our approach, we evaluated it on the PASCAL and NYUD-v2 datasets and compared it against single-task models, standard MTL baselines, and state-of-the-art methods. On the PASCAL benchmark, RobuMTL delivers a +2.8% average relative improvement under single perturbations and up to +44.4% under mixed weather conditions compared to the MTL baseline. On NYUD-v2, RobuMTL achieves a +9.7% average relative improvement across tasks. The code is available at GitHub.
format Preprint
id arxiv_https___arxiv_org_abs_2601_10921
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle RobuMTL: Enhancing Multi-Task Learning Robustness Against Weather Conditions
Shaffee, Tasneem
Reda, Sherief
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Robust Multi-Task Learning (MTL) is crucial for autonomous systems operating in real-world environments, where adverse weather conditions can severely degrade model performance and reliability. In this paper, we introduce RobuMTL, a novel architecture designed to adaptively address visual degradation by dynamically selecting task-specific hierarchical Low-Rank Adaptation (LoRA) modules and a LoRA expert squad based on input perturbations in a mixture-of-experts fashion. Our framework enables adaptive specialization based on input characteristics, improving robustness across diverse real-world conditions. To validate our approach, we evaluated it on the PASCAL and NYUD-v2 datasets and compared it against single-task models, standard MTL baselines, and state-of-the-art methods. On the PASCAL benchmark, RobuMTL delivers a +2.8% average relative improvement under single perturbations and up to +44.4% under mixed weather conditions compared to the MTL baseline. On NYUD-v2, RobuMTL achieves a +9.7% average relative improvement across tasks. The code is available at GitHub.
title RobuMTL: Enhancing Multi-Task Learning Robustness Against Weather Conditions
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2601.10921