Native unified modelling is position as a promising path towards systems that perceive, reason and create within a fully end-to-end framework through SenseNova-U1.5, an 8B-MoT native unified multimodal model that understands, reasons about, and generates visual content within an encoder-free and VAE-free architecture.
Hai-Wen Diao, Jia-Hao Wang, Chen-Jing Ding et al.· 3 citations
This work introduces VBVR-Pro, a closed-loop testbed that makes native visual reasoning through generation trainable, verifiable, optimizable, and experimentally controllable, and identifies recurring failure modes of the prevalent VLM-as-a-judge paradigm.
Junhua Xu, Rui-Si Wang, Fanyi Pu et al.· 5 citations
PhysVICL-74 is introduced, a framework that decomposes physically grounded VICL into physical-rule induction and transition-aligned rendering and TransPhy, a framework that decomposes physically grounded VICL into physical-rule induction and transition-aligned rendering.
Siyi Xie, Xuanke Shi, Jin-Sheng Quan et al.· 0 citations
Experiments show that a single unified model can match leading task-specialized systems across structured visual understanding, dense geometric prediction, segmentation, and multi-view visual geometry.
Xiaoyang Han, Jianhua Li, Kewang Deng et al.· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
