Resolving Conflicts in Lifelong Learning via Aligning Updates in Subspaces

TL;DR

This paper introduces PS-LoRA, a novel framework that aligns updates in the optimization subspace to mitigate catastrophic forgetting in lifelong learning, demonstrating superior performance on NLP and vision tasks.

Contribution

We propose PS-LoRA, a dual-regularization method that aligns task updates and merges adapters, enhancing continual learning stability and efficiency.

Findings

PS-LoRA outperforms existing methods on NLP and vision benchmarks.

It effectively reduces catastrophic forgetting by aligning updates.

The merging strategy consolidates adapters without retraining.

Abstract

Low-Rank Adaptation (LoRA) enables efficient Continual Learning but often suffers from catastrophic forgetting due to destructive interference between tasks. Our analysis reveals that this degradation is primarily driven by antagonistic directional updates where new task gradients directly oppose the historical weight trajectory. To address this, we propose PS-LoRA (Parameter Stability LoRA), a framework designed to resolve conflicts by aligning updates within the optimization subspace. Our approach employs a dual-regularization objective that penalizes conflicting directions and constrains magnitude deviations to ensure consistency with prior knowledge. Additionally, we implement a magnitude-based merging strategy to consolidate sequential adapters into a robust representation without retraining. Experiments on NLP and Vision benchmarks show that PS-LoRA outperforms state-of-the-art methods by preserving the stability of learned representations while efficiently adapting to new domains.