Metraipolis › archive › dc_507ced36f63e
DVD: Dynamic Vector Decoding for Efficient MLLM-based Perception — arXiv:2610.12266
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.12266
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T06:03:51.203Z
- SHA-256 of the archived text
fbc71c1db150a7e97c240c06978f28ef7ef25f077561c8ba166a0fb6c1967b22- Stable URL of this page
https://metraipolis.com/documento/dc_507ced36f63e
How to cite
DVD: Dynamic Vector Decoding for Efficient MLLM-based Perception — arXiv:2610.12266 — arXiv (en), https://arxiv.org/abs/2610.12266 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_507ced36f63e (Metraipolis archive dc_507ced36f63e, sha256 fbc71c1db150a7e97c240c06978f28ef7ef25f077561c8ba166a0fb6c1967b22)
BibTeX
@misc{dc507ced36f63e,
title = {DVD: Dynamic Vector Decoding for Efficient MLLM-based Perception — arXiv:2610.12266},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.12266},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_507ced36f63e; sha256 fbc71c1db150a7e97c240c06978f28ef7ef25f077561c8ba166a0fb6c1967b22},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.12266. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multimodal large language models have made remarkable progress in bridging vision and language, facilitating various perception tasks essential for human-machine interaction, robotics, and autonomous driving. However, existing MLLM-based perception methods predominantly rely on text-based coordinate representation, which suffers from excessive token overhead, or fixed-range quantization, which suffers from range and precision constraints, especially for 3D domains with unbounded spatial range and high localization accuracy requirements. To address these challenges, we propose a dynamic vector decoding method named DVD, which unifies the representation of 2D and 3D perception tasks. Specifically, we first transform diverse perceptual representation (i.e., 2D bounding boxes, 2D masks, and 3D bounding boxes) into 1D vector sequences, which are then mapped to compact discrete tokens in the high-dimensional space. Then, a lightweight de-tokenizer enables seamless integration with MLLMs by decoding output tokens back to original 2D and 3D perceptual representations. Extensive experiments on 2D and 3D perception benchmarks including RefCOCO series, SUN-RGBD, KITTI, Hypersim, nuScenes demonstrate that DVD achieves superior performance in 2D and 3D tasks and reduces significantly the token overhead and inference latency. DVD provides an efficient and general framework for integrating perception capabilities into MLLMs, overcoming the inherent limitations of existing methods. 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/2610.12266 — 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 2026-10-09T05:47:02.114Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).