6 papers
TerraVis: Towards Evaluation of World-Grounded Visual Consistency in Text-to-Image Generation via MLLM Workflows
Shuai Fu, Jing Gu, Jian Zhou +3
Recent text-to-image models have made substantial progress in photorealism, aesthetics, and text-image alignment. Yet visually appealing images can still violate real-world plausib…
Embodied Agents Take Control: Minimal-Interface Zero-Shot Agents Rival Industrial-Scale Policies in Vision-and-Language Navigation
Jian Zhou, Xunyi Zhao, Gengze Zhou +4
Autonomous embodied agents must sustain a long decision-making loop of perceiving, acting, verifying, and self-correcting. In trained end-to-end policies and hand-designed workflow…
Automating the Design of Embodied Agent Architectures
Jian Zhou, Sihao Lin, Jin Li +3
Embodied agents are typically built as hand-designed compositions of perception, memory, planning, and action modules. This modularity exposes a large architectural design space, b…
One Agent to Guide Them All: Empowering MLLMs for Vision-and-Language Navigation via Explicit World Representation
Zerui Li, Hongpei Zheng, Fangguo Zhao +5
A navigable agent needs to understand both high-level semantic instructions and precise spatial perceptions. Building navigation agents centered on Multimodal Large Language Models…
Freeze, Share, Shrink: Rethinking the Action Backbone in Diffusion Policies
Jian Zhou, Sihao Lin, Shuai Fu +3
Many recent Vision-Language-Action models employ diffusion or flow-matching backbones with hundreds of millions of parameters for action generation. However, unlike image synthesis…
Counting Hallucinations in Diffusion Models
Shuai Fu, Jian Zhou, Qi Chen +7
Diffusion probabilistic models (DPMs) have demonstrated remarkable progress in generative tasks, such as image and video synthesis. However, they still often produce hallucinated s…