8 papers
Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs
Yunheng Li, Guohong Mu, Hao Li +4
Multimodal large language models (MLLMs) have become a prevailing paradigm for unified video perception. However, post-training on large multi-task datasets remains challenging, as…
PhysMani: Physics-principled 3D World Model for Dynamic Object Manipulation
Peng Yun, Shouwang Huang, Hao Li +3
Manipulating fast and dynamically moving targets in unstructured 3D environments remains challenging for embodied AI. Existing visual-language-action models and world models strugg…
Golden RPG: Confidence-Adaptive Region-Aware Noise for Compositional Text-to-Image Generation
Hao Li
Compositional text-to-image (T2I) generation requires a model to honour multiple sub-prompts that describe distinct image regions. Recent work shows that the \emph{starting noise}…
PhysInOne: Visual Physics Learning and Reasoning in One Suite
Siyuan Zhou, Hejun Wang, Hu Cheng +36
We present PhysInOne, a large-scale synthetic dataset addressing the critical scarcity of physically-grounded training data for AI systems. Unlike existing datasets limited to mere…
Uni-ViGU: Towards Unified Video Generation and Understanding via A Diffusion-Based Video Generator
Luozheng Qin, Jia Gong, Qian Qiao +6
Unified multimodal models integrating visual understanding and generation face a fundamental challenge: visual generation incurs substantially higher computational costs than under…
Steering Video Diffusion Transformers with Massive Activations
Xianhang Cheng, Yujian Zheng, Zhenyu Xie +2
In this work, we study the role of Massive Activations (MAs), which are rare, high-magnitude spikes confined to a few fixed hidden dimensions in video diffusion transformers (DiTs)…
T2I-R1: Reinforcing Image Generation with Collaborative Semantic-level and Token-level CoT
Dongzhi Jiang, Ziyu Guo, Renrui Zhang +6
Recent advancements in large language models have demonstrated how chain-of-thought (CoT) and reinforcement learning (RL) can improve performance. However, applying such reasoning…
Adaptive Markup Language Generation for Contextually-Grounded Visual Document Understanding
Han Xiao, Yina Xie, Guanxin Tan +12
Visual Document Understanding has become essential with the increase of text-rich visual content. This field poses significant challenges due to the need for effective integration…