Metraipolis › archive › dc_663a4380cfcb
ResComEmb: Effective and Efficient Multimodal Embedding via Residual Homogeneity Compression — arXiv:2609.37225
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.37225
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T16:27:02.175Z
- SHA-256 of the archived text
25e7bacee699be093ae67e7c529279c212ac228a706d400f745290c86bd7f319- Stable URL of this page
https://metraipolis.com/documento/dc_663a4380cfcb
How to cite
ResComEmb: Effective and Efficient Multimodal Embedding via Residual Homogeneity Compression — arXiv:2609.37225 — arXiv (en), https://arxiv.org/abs/2609.37225 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_663a4380cfcb (Metraipolis archive dc_663a4380cfcb, sha256 25e7bacee699be093ae67e7c529279c212ac228a706d400f745290c86bd7f319)
BibTeX
@misc{dc663a4380cfcb,
title = {ResComEmb: Effective and Efficient Multimodal Embedding via Residual Homogeneity Compression — arXiv:2609.37225},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.37225},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_663a4380cfcb; sha256 25e7bacee699be093ae67e7c529279c212ac228a706d400f745290c86bd7f319},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37225. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multimodal large language models (MLLMs) have shown strong potential for universal multimodal representation learning. However, existing methods either compress each input into a single vector, limiting fine-grained expressiveness, or retain long sequences of visual-token vectors, incurring substantial storage and interaction costs. To resolve this trade-off, we propose ResComEmb, a trainable framework for effective and efficient universal multi-vector multimodal embedding. ResComEmb first encodes each input at native dynamic resolution into ordered global, intermediate, and fine-grained views. After MLLM contextualization and embedding projection, a trainable Residual Homogeneity Compression (RHC) module reduces within-granularity redundancy and cross-granularity repetition under explicit visual token budgets. Then, ResComEmb introduces a length-adaptive Bidirectional Late-Interaction Matching mechanism for robust query-document scoring, which averages the strongest token-level matches in each direction and combines the two scores using a weight based on how many valid tokens each side has. Extensive experiments on MMEB, ViDoRe V1, and ViDoRe V2 show that ResComEmb produces higher-quality universal multimodal embeddings than VLM2Vec-V2, and outperforms ColQwen2.5 in visual document retrieval using only 37.5% of its full visual token budget, demonstrating a favorable effectiveness-efficiency trade-off. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2609.37225 — 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 2026-10-07T15:54:53.958Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).