Research

Paper

AI LLM March 02, 2026

DeLo: Dual Decomposed Low-Rank Experts Collaboration for Continual Missing Modality Learning

Authors

Xiwei Liu, Yulong Li, Feilong Tang, Imran Razzak

Abstract

Adapting Large Multimodal Models (LMMs) to real-world scenarios poses the dual challenges of learning from sequential data streams while handling frequent modality incompleteness, a task known as Continual Missing Modality Learning (CMML). However, existing works on CMML have predominantly relied on prompt tuning, a technique that struggles with this task due to cross-task interference between its learnable prompts in their shared embedding space. A naive application of Low-Rank Adaptation (LoRA) with modality-shared module will also suffer modality interference from competing gradients. To this end, we propose DeLo, the first framework to leverage a novel dual-decomposed low-rank expert architecture for CMML. Specifically, this architecture resolves modality interference through decomposed LoRA expert, dynamically composing LoRA update matrix with rank-one factors from disentangled modality-specific factor pools. Embedded within a task-partitioned framework that structurally prevents catastrophic forgetting, this expert system is supported by two key mechanisms: a Cross-Modal Guided Routing strategy to handle incomplete data and a Task-Key Memory for efficient, task-agnostic inference. Extensive experiments on established CMML benchmarks demonstrate that our method significantly outperforms state-of-the-art approaches. This highlights the value of a principled, architecturally-aware LoRA design for real-world multimodal challenges.

Metadata

arXiv ID: 2603.01632
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-02
Fetched: 2026-03-03 04:34

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.01632v1</id>\n    <title>DeLo: Dual Decomposed Low-Rank Experts Collaboration for Continual Missing Modality Learning</title>\n    <updated>2026-03-02T09:07:28Z</updated>\n    <link href='https://arxiv.org/abs/2603.01632v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.01632v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Adapting Large Multimodal Models (LMMs) to real-world scenarios poses the dual challenges of learning from sequential data streams while handling frequent modality incompleteness, a task known as Continual Missing Modality Learning (CMML). However, existing works on CMML have predominantly relied on prompt tuning, a technique that struggles with this task due to cross-task interference between its learnable prompts in their shared embedding space. A naive application of Low-Rank Adaptation (LoRA) with modality-shared module will also suffer modality interference from competing gradients. To this end, we propose DeLo, the first framework to leverage a novel dual-decomposed low-rank expert architecture for CMML. Specifically, this architecture resolves modality interference through decomposed LoRA expert, dynamically composing LoRA update matrix with rank-one factors from disentangled modality-specific factor pools. Embedded within a task-partitioned framework that structurally prevents catastrophic forgetting, this expert system is supported by two key mechanisms: a Cross-Modal Guided Routing strategy to handle incomplete data and a Task-Key Memory for efficient, task-agnostic inference. Extensive experiments on established CMML benchmarks demonstrate that our method significantly outperforms state-of-the-art approaches. This highlights the value of a principled, architecturally-aware LoRA design for real-world multimodal challenges.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-02T09:07:28Z</published>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Xiwei Liu</name>\n    </author>\n    <author>\n      <name>Yulong Li</name>\n    </author>\n    <author>\n      <name>Feilong Tang</name>\n    </author>\n    <author>\n      <name>Imran Razzak</name>\n    </author>\n  </entry>"
}