Continual-NExT: A Unified Comprehension And Generation Continual Learning Framework
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908843311628288 |
|---|---|
| author | Qiao, Jingyang Zhang, Zhizhong Tan, Xin Gong, Jingyu Qu, Yanyun Xie, Yuan |
| author_facet | Qiao, Jingyang Zhang, Zhizhong Tan, Xin Gong, Jingyu Qu, Yanyun Xie, Yuan |
| contents | Dual-to-Dual MLLMs refer to Multimodal Large Language Models, which can enable unified multimodal comprehension and generation through text and image modalities. Although exhibiting strong instantaneous learning and generalization capabilities, Dual-to-Dual MLLMs still remain deficient in lifelong evolution, significantly affecting continual adaptation to dynamic real-world scenarios. One of the challenges is that learning new tasks inevitably destroys the learned knowledge. Beyond traditional catastrophic forgetting, Dual-to-Dual MLLMs face other challenges, including hallucination, instruction unfollowing, and failures in cross-modal knowledge transfer. However, no standardized continual learning framework for Dual-to-Dual MLLMs has been established yet, leaving these challenges unexplored. Thus, in this paper, we establish Continual-NExT, a continual learning framework for Dual-to-Dual MLLMs with deliberately-architected evaluation metrics. To improve the continual learning capability of Dual-to-Dual MLLMs, we propose an efficient MAGE (Mixture and Aggregation of General LoRA and Expert LoRA) method to further facilitate knowledge transfer across modalities and mitigate forgetting. Extensive experiments demonstrate that MAGE outperforms other continual learning methods and achieves state-of-the-art performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_18055 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Continual-NExT: A Unified Comprehension And Generation Continual Learning Framework Qiao, Jingyang Zhang, Zhizhong Tan, Xin Gong, Jingyu Qu, Yanyun Xie, Yuan Machine Learning Dual-to-Dual MLLMs refer to Multimodal Large Language Models, which can enable unified multimodal comprehension and generation through text and image modalities. Although exhibiting strong instantaneous learning and generalization capabilities, Dual-to-Dual MLLMs still remain deficient in lifelong evolution, significantly affecting continual adaptation to dynamic real-world scenarios. One of the challenges is that learning new tasks inevitably destroys the learned knowledge. Beyond traditional catastrophic forgetting, Dual-to-Dual MLLMs face other challenges, including hallucination, instruction unfollowing, and failures in cross-modal knowledge transfer. However, no standardized continual learning framework for Dual-to-Dual MLLMs has been established yet, leaving these challenges unexplored. Thus, in this paper, we establish Continual-NExT, a continual learning framework for Dual-to-Dual MLLMs with deliberately-architected evaluation metrics. To improve the continual learning capability of Dual-to-Dual MLLMs, we propose an efficient MAGE (Mixture and Aggregation of General LoRA and Expert LoRA) method to further facilitate knowledge transfer across modalities and mitigate forgetting. Extensive experiments demonstrate that MAGE outperforms other continual learning methods and achieves state-of-the-art performance. |
| title | Continual-NExT: A Unified Comprehension And Generation Continual Learning Framework |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.18055 |