5 papers
DM0: An Embodied-Native Vision-Language-Action Model towards Physical AI
En Yu, Haoran Lv, Jianjian Sun +46
Moving beyond the traditional paradigm of adapting internet-pretrained models to physical tasks, we present DM0, an Embodied-Native Vision-Language-Action (VLA) framework designed…
SceneFoundry: Generating Interactive Infinite 3D Worlds
ChunTeng Chen, YiChen Hsu, YiWen Liu +5
The ability to automatically generate large-scale, interactive, and physically realistic 3D environments is crucial for advancing robotic learning and embodied intelligence. Howeve…
Beyond Masked and Unmasked: Discrete Diffusion Models via Partial Masking
Chen-Hao Chao, Wei-Fang Sun, Hanwen Liang +2
Masked diffusion models (MDM) are powerful generative models for discrete data that generate samples by progressively unmasking tokens in a sequence. Each token can take one of two…
EDELINE: Enhancing Memory in Diffusion-based World Models via Linear-Time Sequence Modeling
Jia-Hua Lee, Bor-Jiun Lin, Wei-Fang Sun +1
World models represent a promising approach for training reinforcement learning agents with significantly improved sample efficiency. While most world model methods primarily rely…
Retraining-Free Merging of Sparse MoE via Hierarchical Clustering
I-Chun Chen, Hsu-Shen Liu, Wei-Fang Sun +3
Sparse Mixture-of-Experts (SMoE) models represent a significant advancement in large language model (LLM) development through their efficient parameter utilization. These models ac…