32 citations · 63 across the 11 of their papers we have counts for
14 papers
Can Video World Models Track Unobserved World States?
Joonghyuk Shin, Yicong Hong, Jaesik Park +1
Video world models are increasingly used as simulators, but visual fidelity alone does not show that a model maintains the hidden state of the world. We examine this difference wit…
Chimera: Designing and Chinchilla-Scaling Hybrid Visual Diffusion Transformers
Chongjian Ge, Hanwen Jiang, Tianyu Wang +9
Visual generation increasingly requires high-resolution images, long videos, and multimodal context, making the quadratic cost of full attention prohibitive. We introduce Chimera,…
WorldCam: Interactive Autoregressive 3D Gaming Worlds with Camera Pose as a Unifying Geometric Representation
Jisu Nam, Yicong Hong, Chun-Hao Paul Huang +9
Recent advances in video diffusion transformers have enabled interactive gaming world models that allow users to explore generated environments over extended horizons. However, exi…
Coarse-to-Real: Generative Rendering for Populated Dynamic Scenes
Gonzalo Gomez-Nogales, Yicong Hong, Chongjian Ge +4
Traditional rendering pipelines rely on complex assets, accurate materials and lighting, and substantial computational resources to produce realistic imagery, yet they still face c…
Rethinking Training Dynamics in Scale-wise Autoregressive Generation
Gengze Zhou, Chongjian Ge, Hao Tan +2
Recent advances in autoregressive (AR) generative models have produced increasingly powerful systems for media synthesis. Among them, next-scale prediction has emerged as a popular…
RELIC: Interactive Video World Model with Long-Horizon Memory
Yicong Hong, Yiqun Mei, Chongjian Ge +11
A truly interactive world model requires three key ingredients: real-time long-horizon streaming, consistent spatial memory, and precise user control. However, most existing approa…
Diffusion Transformer-to-Mamba Distillation for High-Resolution Image Generation
Yuan Yao, Yicong Hong, Difan Liu +3
The quadratic computational complexity of self-attention in diffusion transformers (DiT) introduces substantial computational costs in high-resolution image generation. While the l…
Test-Time Training Done Right
Tianyuan Zhang, Sai Bi, Yicong Hong +6
Test-Time Training (TTT) models context dependencies by adapting part of the model's weights (referred to as fast weights) during inference. This fast weight, akin to recurrent sta…
REGEN: Learning Compact Video Embedding with (Re-)Generative Decoder
Yitian Zhang, Long Mai, Aniruddha Mahapatra +5
We present a novel perspective on learning video embedders for generative modeling: rather than requiring an exact reproduction of an input video, an effective embedder should focu…
Pushing the Boundaries of State Space Models for Image and Video Generation
Yicong Hong, Long Mai, Yuan Yao +1
While Transformers have become the dominant architecture for visual generation, linear attention models, such as the state-space models (SSM), are increasingly recognized for their…
Long-LRM: Long-sequence Large Reconstruction Model for Wide-coverage Gaussian Splats
Chen Ziwen, Hao Tan, Kai Zhang +5
We propose Long-LRM, a feed-forward 3D Gaussian reconstruction model for instant, high-resolution, 360° wide-coverage, scene-level reconstruction. Specifically, it takes in 32 inpu…
Progressive Autoregressive Video Diffusion Models
Desai Xie, Zhan Xu, Yicong Hong +5
Current frontier video diffusion models have demonstrated remarkable results at generating high-quality videos. However, they can only generate short video clips, normally around 1…
Instant3D: Fast Text-to-3D with Sparse-View Generation and Large Reconstruction Model
Jiahao Li, Hao Tan, Kai Zhang +7
Text-to-3D with diffusion models has achieved remarkable progress in recent years. However, existing methods either rely on score distillation-based optimization which suffer from…
LRM: Large Reconstruction Model for Single Image to 3D
Yicong Hong, Kai Zhang, Jiuxiang Gu +7
We propose the first Large Reconstruction Model (LRM) that predicts the 3D model of an object from a single input image within just 5 seconds. In contrast to many previous methods…