Jiatao Gu

Affiliations

Facebook AI Research

papers

3,997

total citations

papers (14)

Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstruction

ICCV 2023arXiv

199

citations

Diffusion Models Without Attention

CVPR 2024arXiv

citations

Cross-lingual Retrieval for Iterative Self-Supervised Training

NEURIPS 2020arXiv

citations

Normalizing Flows are Capable Generative Models

ICML 2025arXiv

citations

PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Model

NEURIPS 2023arXiv

citations

World-consistent Video Diffusion with Explicit 3D Modeling

CVPR 2025arXiv

citations

Denoising Autoregressive Transformers for Scalable Text-to-Image Generation

ICLR 2025arXiv

citations

Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation

CVPR 2025arXiv

citations

Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows

NEURIPS 2025arXiv

citations

TADA: Improved Diffusion Sampling with Training-free Augmented DynAmics

NEURIPS 2025arXiv

citations

Data-free Distillation of Diffusion Models with Bootstrapping

ICML 2024

citations

Jiatao Gu

Affiliations

papers (14)

Neural Sparse Voxel Fields

Volume Rendering of Neural Implicit Surfaces

PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding

Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstruction

Diffusion Models Without Attention

Cross-lingual Retrieval for Iterative Self-Supervised Training

Normalizing Flows are Capable Generative Models

PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Model

World-consistent Video Diffusion with Explicit 3D Modeling

Denoising Autoregressive Transformers for Scalable Text-to-Image Generation

Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation

Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows

TADA: Improved Diffusion Sampling with Training-free Augmented DynAmics

Data-free Distillation of Diffusion Models with Bootstrapping

papers (14)

Neural Sparse Voxel Fields

Volume Rendering of Neural Implicit Surfaces

PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding

Single-Stage Diffusion NeRF: A Unified Approach to 3D Generation and Reconstruction

Diffusion Models Without Attention

Cross-lingual Retrieval for Iterative Self-Supervised Training

Normalizing Flows are Capable Generative Models

PLANNER: Generating Diversified Paragraph via Latent Language Diffusion Model

World-consistent Video Diffusion with Explicit 3D Modeling

Denoising Autoregressive Transformers for Scalable Text-to-Image Generation

Vid2Sim: Generalizable, Video-based Reconstruction of Appearance, Geometry and Physics for Mesh-free Simulation

Flexible Language Modeling in Continuous Space with Transformer-based Autoregressive Flows

TADA: Improved Diffusion Sampling with Training-free Augmented DynAmics

Data-free Distillation of Diffusion Models with Bootstrapping