Metraipolis › archive › dc_7dfc5ccefb9d

COMET: Decoupled Distillation, Routing, and Capacity Control for Task-Agnostic Continual Vision--Language Learning — arXiv:2610.22907

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.22907
License (as archived)
CC0-1.0
Archived
2026-10-19T08:43:03.366Z
SHA-256 of the archived text
ab2126b86e497b75acca04c847d203291100decf7a6273754b46c1ceba5939cb
Stable URL of this page
https://metraipolis.com/documento/dc_7dfc5ccefb9d

How to cite

COMET: Decoupled Distillation, Routing, and Capacity Control for Task-Agnostic Continual Vision--Language Learning — arXiv:2610.22907 — arXiv (en), https://arxiv.org/abs/2610.22907 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7dfc5ccefb9d (Metraipolis archive dc_7dfc5ccefb9d, sha256 ab2126b86e497b75acca04c847d203291100decf7a6273754b46c1ceba5939cb)
BibTeX
@misc{dc7dfc5ccefb9d,
  title = {COMET: Decoupled Distillation, Routing, and Capacity Control for Task-Agnostic Continual Vision--Language Learning — arXiv:2610.22907},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.22907},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7dfc5ccefb9d; sha256 ab2126b86e497b75acca04c847d203291100decf7a6273754b46c1ceba5939cb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22907. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Vision-language models (VLMs) such as CLIP retain zero-shot recognition ability, but their open-world use often requires task-agnostic continual learning: tasks arrive sequentially, past data are unavailable, task identity is unknown, and the backbone cannot be retrained. We argue that this setting fails VLMs through three coupled pressures beyond the stability-plasticity tradeoff: (1) modal-consistency drift, (2) representation-space interference, and (3) per-task capacity shortfall. We propose COMET (Collaborative Mixture-of-Expert Transfer) to resolve these three issues respectively as follows. A frozen teacher provides only a feature reference on a shared image-text pool; Tri-Affinity Distillation preserves image-image, text-text, and image-text geometry; and Student-Aware Cross-Expert Distillation aligns new experts with compatible prior experts so routing errors degrade gracefully. We design COMET to keep capacity and inference separate: a contextual bandit keeps or merges full per-task experts using post-training signals, while a reconstruction router selects sparse experts at test time without teacher semantics. On the X-TAIL benchmark, we show that each proposed component improves continual learning performance. COMET shows that continual vision-language adaptation can preserve multimodal geometry, coordinate accumulated experts, and allocate capacity jointly when feature reference, expert shaping, routing, and capacity control are separated by construction.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.22907 — 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 2026-10-19T08:39:10.134Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).