North Carolina State University(北卡罗来纳州立大学);UC, San Diego(加州大学圣地亚哥分校);University of Alabama at Birmingham(阿拉巴马大学伯明翰分校);Johns Hopkins University(约翰霍普金斯大学);Duke University(杜克大学);Boston University(波士顿大学);The Ohio State University(俄亥俄州立大学)
Dongyao Zhu, Zhen Wang, Xi Xiao, Han Jiang, Saeed Vahidian, Wei-Lun Chao, Tanya Berger-Wolf, Yu Su, Raju Vatsavai, Jianyang Gu
Latent visual reasoning involves visual evidence more directly in multimodal reasoning by inserting continuous latent tokens before textual generation. However, the necessity of these latent tokens at inference remains ambiguous. We show that replacing latent tokens with random noise or removing them completely causes little performance degradation across spatial reasoning benchmarks. Reinforcement learning further diminishes the latent generation behavior after post-training. These observations raise a central question: \textit{Is latent visual reasoning still meaningful}? We argue that its value should be measured by how effectively latent tokens guide learning, rather than whether they persist as an inference-time format. Our analysis shows that latent reasoning is unevenly favorable across question types, yet hard task-level routing for applying latent generation is brittle. Motivated by these findings, we propose an attention-based reward that encourages generated latent tokens to interact with later text tokens during RL. This reward promotes latent utilization when the latent mode is activated while preserving the flexibility to use pure-text reasoning. Experiments show that our method improves performance across perception and visual reasoning benchmarks, even when latent tokens are rarely generated after post-training. Our results highlight that, without explicit expression at inference, latent visual reasoning can shape better visual grounding and more accurate textual reasoning \textbf{in silence}. Code \& model: \href{https://github.com/ddydyd32/silent-lvr/tree/master}{\faGithub}.
Not Blind but Silenced: Rebalancing Vision and Language via Adversarial Counter-Commonsense Equilibrium
South China University of Technology(华南理工大学);Institute for Super Robotics (Huangpu)(机器人研究所(黄埔));Shanghai Jiao Tong University(上海交通大学);Changsha University of Science and Technology(长沙理工大学)
Silence is Golden: Leveraging Adversarial Examples to Nullify Audio Control in LDM-based Talking-Head Generation
ReLER, CCAI, Zhejiang University(ReLER,中国人工智能学会,浙江大学);School of Cyber Science and Technology, Sun Yat-sen University(信息科学与技术学院,孙中山大学);Department of Statistics, University of Wisconsin–Madison(统计学系,威斯康星大学麦迪逊分校)
Leveraging Scene Embeddings for Gradient-Based Motion Planning in Latent Space
Applied AI Lab (A2I)(应用人工智能实验室(A2I));Dynamic Robot Systems (DRS)(动态机器人系统实验室(DRS));Oxford Robotics Institute (ORI)(牛津机器人研究所(ORI));University of Oxford(牛津大学)
Make-A-Volume: Leveraging Latent Diffusion Models for Cross-Modality 3D Brain MRI Synthesis
The University of Hong Kong(香港大学);The Chinese University of Hong Kong(香港中文大学);Qilu Hospital of Shandong University(山东大学齐鲁医院);Nanyang Technological University(南洋理工大学)