10 papers
Spatial Memory Intelligence: Endowing World Models with Understanding-Driven Long-Term Memory
Ying Yang, Guiyu Zhang, Lianghua Huang +5
Long-video generation and world models have shown strong potential for interactive entertainment and embodied simulation by predicting future observations conditioned on user actio…
PrecogUI: Proactive GUI Agents via Pre-cognitive Simulation and Experience Retrieval
Bin Kang, Jiarui Ouyang, Li Jiang +2
Existing reactive Graphical User Interface (GUI) agents often fail in long-horizon, dynamic scenarios, where unexpected disturbances trigger attention-diverting and cascading failu…
SolarWM: Open Data and Scalable Training for Long-Horizon Video World Models
Junchao Huang, Guian Fang, Shengju Qian +15
We introduce SolarWM, a fully open foundation for building interactive video world models from data preparation through long-horizon inference. Training across heterogeneous data s…
VideoRAE: Taming Video Foundation Models for Generative Modeling via Representation Autoencoders
Zhihao Xie, Junfeng Wu, Xinting Hu +2
Video generation models typically rely on 3D-VAEs trained for pixel-level reconstruction, whose latent spaces may underrepresent semantic structure. We introduce VideoRAE, a repres…
Denoising Tells When to Replan: Denoising-Variance Adaptive Chunking for Flow-Based Robot Policies
Xiangdong Feng, Yuxuan Cheng, Chen Shi +5
Action chunking has become a common inference strategy for flow-based robot policies, improving action coherence by modeling multi-step temporal dependencies in demonstrations. How…
FlashVID: Efficient Video Large Language Models via Training-free Tree-based Spatiotemporal Token Merging
Ziyang Fan, Keyu Chen, Ruilong Xing +3
Although Video Large Language Models (VLLMs) have shown remarkable capabilities in video understanding, they are required to process high volumes of visual tokens, causing signific…
Memory Forcing: Spatio-Temporal Memory for Consistent Scene Generation on Minecraft
Junchao Huang, Xinting Hu, Boyao Han +4
Autoregressive video diffusion models have proved effective for world modeling and interactive scene generation, with Minecraft gameplay as a representative application. To faithfu…
Mitigating Object Hallucinations via Sentence-Level Early Intervention
Shangpin Peng, Senqiao Yang, Li Jiang +1
Multimodal large language models (MLLMs) have revolutionized cross-modal understanding but continue to struggle with hallucinations - fabricated content contradicting visual inputs…
Edit360: 2D Image Edits to 3D Assets from Any Angle
Junchao Huang, Xinting Hu, Shaoshuai Shi +2
Recent advances in diffusion models have significantly improved image generation and editing, but extending these capabilities to 3D assets remains challenging, especially for fine…
Unified Language-driven Zero-shot Domain Adaptation
Senqiao Yang, Zhuotao Tian, Li Jiang +1
This paper introduces Unified Language-driven Zero-shot Domain Adaptation (ULDA), a novel task setting that enables a single model to adapt to diverse target domains without explic…