Metraipolis › archive › dc_a6458bd76a17

CoRe-MoE: Compact Reusable MoE for Continual Multimodal Instruction Tuning — arXiv:2608.27867

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.27867
License (as archived)
CC0-1.0
Archived
2026-10-17T10:47:16.172Z
SHA-256 of the archived text
43e81c1f006ef8cd782c2f882b4214f3a464cb96f627ef71de35f4e809427d61
Stable URL of this page
https://metraipolis.com/documento/dc_a6458bd76a17

How to cite

CoRe-MoE: Compact Reusable MoE for Continual Multimodal Instruction Tuning — arXiv:2608.27867 — arXiv (en), https://arxiv.org/abs/2608.27867 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a6458bd76a17 (Metraipolis archive dc_a6458bd76a17, sha256 43e81c1f006ef8cd782c2f882b4214f3a464cb96f627ef71de35f4e809427d61)
BibTeX
@misc{dca6458bd76a17,
  title = {CoRe-MoE: Compact Reusable MoE for Continual Multimodal Instruction Tuning — arXiv:2608.27867},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.27867},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a6458bd76a17; sha256 43e81c1f006ef8cd782c2f882b4214f3a464cb96f627ef71de35f4e809427d61},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27867. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Continual multimodal instruction tuning requires multimodal large language models to acquire new task abilities sequentially while preserving previously learned knowledge. LoRA-MoE provides a promising solution by introducing expert-based capacity, but repeatedly learning and maintaining full LoRA experts leads to substantial parameter overhead. This raises a natural question: is full expert expansion necessary for every new task? To answer it, we analyze the SVD of task-specific LoRA updates and observe substantial overlap in their input- and output-side LoRA direction subspaces, with task-specific adaptation largely captured by lightweight coordinates over these subspaces. Motivated by this observation, we propose CoRe-MoE, a Compact Reusable MoE framework for parameter-efficient continual multimodal instruction tuning. CoRe-MoE extracts reusable input- and output-side direction bases from an initial expert bank, and for subsequent tasks trains only compact coordinate experts together with task-specific low-rank routers. Experiments on two representative MLLMs show that CoRe-MoE improves final average performance over the strongest competing baseline by up to 5.90 points, while using less than 1% of the trainable parameters required by sequential LoRA for later tasks. The code is publicly available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.27867 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T10:33:42.360Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).