Decomposing and Composing: Towards Efficient Vision-Language Continual Learning via Rank-1 Expert Pool in a Single LoRA

TL;DR

This paper introduces a novel vision-language continual learning framework using a Rank-1 Expert Pool within a single LoRA module, enabling efficient, domain-aware task adaptation with minimal parameters and no external knowledge reliance.

Contribution

The paper proposes a decomposable Rank-1 Expert Pool within a single LoRA, combined with an Activation-Guided Orthogonal loss, to improve continual learning efficiency and performance in vision-language models.

Findings

Achieves state-of-the-art results across multiple benchmarks.

Reduces trainable parameters by 96.7% compared to baseline.

Eliminates inference latency and external dataset reliance.

Abstract

Continual learning (CL) in vision-language models (VLMs) faces significant challenges in improving task adaptation and avoiding catastrophic forgetting. Existing methods usually have heavy inference burden or rely on external knowledge, while Low-Rank Adaptation (LoRA) has shown potential in reducing these issues by enabling parameter-efficient tuning. However, considering directly using LoRA to alleviate the catastrophic forgetting problem is non-trivial, we introduce a novel framework that restructures a single LoRA module as a decomposable Rank-1 Expert Pool. Our method learns to dynamically compose a sparse, task-specific update by selecting from this expert pool, guided by the semantics of the [CLS] token. In addition, we propose an Activation-Guided Orthogonal (AGO) loss that orthogonalizes critical parts of LoRA weights across tasks. This sparse composition and orthogonalization enable fewer parameter updates, resulting in domain-aware learning while minimizing inter-task interference and maintaining downstream task performance. Extensive experiments across multiple settings demonstrate state-of-the-art results in all metrics, surpassing zero-shot upper bounds in generalization. Notably, it reduces trainable parameters by 96.7% compared to the baseline method, eliminating reliance on external datasets or task-ID discriminators. The merged LoRAs retain less weights and incur no inference latency, making our method computationally lightweight.