Rethinking Efficient Mixture-of-Experts for Remote Sensing Modality-Missing Classification
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911417357041664 |
|---|---|
| author | Gao, Qinghao Qu, Jiahui Dong, Wenqian |
| author_facet | Gao, Qinghao Qu, Jiahui Dong, Wenqian |
| contents | Multimodal remote sensing classification often suffers from missing modalities caused by sensor failures and environmental interference, leading to severe performance degradation. In this work, we rethink missing-modality learning from a conditional computation perspective and investigate whether Mixture-of-Experts (MoE) models can inherently adapt to diverse modality-missing scenarios. We first conduct a systematic study of representative MoE paradigms under various missing-modality settings, revealing both their potential and limitations. Building on these insights, we propose a Missing-aware Mixture-of-LoRAs (MaMOL), a parameter-efficient MoE framework that unifies multiple modality-missing cases within a single model. MaMOL introduces a dual-routing mechanism to decouple modality-invariant shared experts and modality-aware dynamic experts, enabling automatic expert activation conditioned on available modalities. Extensive experiments on multiple remote sensing benchmarks demonstrate that MaMOL significantly improves robustness and generalization under diverse missing-modality scenarios with minimal computational overhead. Transfer experiments on natural image datasets further validate its scalability and cross-domain applicability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_11460 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Rethinking Efficient Mixture-of-Experts for Remote Sensing Modality-Missing Classification Gao, Qinghao Qu, Jiahui Dong, Wenqian Computer Vision and Pattern Recognition Multimodal remote sensing classification often suffers from missing modalities caused by sensor failures and environmental interference, leading to severe performance degradation. In this work, we rethink missing-modality learning from a conditional computation perspective and investigate whether Mixture-of-Experts (MoE) models can inherently adapt to diverse modality-missing scenarios. We first conduct a systematic study of representative MoE paradigms under various missing-modality settings, revealing both their potential and limitations. Building on these insights, we propose a Missing-aware Mixture-of-LoRAs (MaMOL), a parameter-efficient MoE framework that unifies multiple modality-missing cases within a single model. MaMOL introduces a dual-routing mechanism to decouple modality-invariant shared experts and modality-aware dynamic experts, enabling automatic expert activation conditioned on available modalities. Extensive experiments on multiple remote sensing benchmarks demonstrate that MaMOL significantly improves robustness and generalization under diverse missing-modality scenarios with minimal computational overhead. Transfer experiments on natural image datasets further validate its scalability and cross-domain applicability. |
| title | Rethinking Efficient Mixture-of-Experts for Remote Sensing Modality-Missing Classification |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.11460 |