8 papers
ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training
Haian Jin, Rundi Wu, Tianyuan Zhang +4
Feed-forward transformer models have driven rapid progress in 3D vision, but state-of-the-art methods such as VGGT and have a computational cost that scales quadratically wit…
AlignTok: Aligning Visual Foundation Encoders to Tokenizers for Diffusion Models
Bowei Chen, Sai Bi, Hao Tan +6
In this work, we propose aligning pretrained visual encoders to serve as tokenizers for latent diffusion models in image generation. Unlike training a variational autoencoder (VAE)…
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…
REG: Rectified Gradient Guidance for Conditional Diffusion Models
Zhengqi Gao, Kaiwen Zha, Tianyuan Zhang +2
Guidance techniques are simple yet effective for improving conditional generation in diffusion models. Albeit their empirical success, the practical implementation of guidance dive…
RandAR: Decoder-only Autoregressive Visual Generation in Random Orders
Ziqi Pang, Tianyuan Zhang, Fujun Luan +5
We introduce RandAR, a decoder-only visual autoregressive (AR) model capable of generating images in arbitrary token orders. Unlike previous decoder-only AR models that rely on a p…
Buffer Anytime: Zero-Shot Video Depth and Normal from Image Priors
Zhengfei Kuang, Tianyuan Zhang, Kai Zhang +7
We present Buffer Anytime, a framework for estimation of depth and normal maps (which we call geometric buffers) from video that eliminates the need for paired video--depth and vid…