Hao Tan 「谭昊」

I am a Senior Research Scientist at Adobe Research. I work on generative models for virtual worlds and the real world, spanning long-context memory, 3D and video generation, and multimodal learning.

I received my Ph.D. from UNC Chapel Hill, advised by Mohit Bansal, and my B.S. from Shanghai Jiao Tong University's ACM Honors Class.

Selected Publications

For a complete publication list, see Google Scholar.

Chimera: Designing and Chinchilla-Scaling Hybrid Visual Diffusion Transformers

Chongjian Ge^*, Hanwen Jiang^*, Tianyu Wang^*, Jiuxiang Gu, Yiran Xu, Ziwen Chen, Shaoteng Liu, Jing Shi, Yicong Hong, Zefan Cai, Hailin Jin, Hao Tan

arXiv, 2026

paper

Hybrid KDA-MLA diffusion Transformers with sparse MoE layers and HeteroP scaling laws for efficient long-context image and video generation.

RELIC: Interactive Video World Model with Long-Horizon Memory

Yicong Hong^*, Yiqun Mei^*, Chongjian Ge^*, Yiran Xu, Yang Zhou, Sai Bi, Yannick Hold-Geoffroy, Mike Roberts, Matthew Fisher, Eli Shechtman, Kalyan Sunkavalli, Feng Liu, Zhengqi Li, Hao Tan

Tech report, 2025

project page / paper

Real-time interactive video generation with compact, camera-aware long-horizon memory.

Test-Time Training Done Right

Tianyuan Zhang, Sai Bi, Yicong Hong, Kai Zhang, Fujun Luan, Songlin Yang, Kalyan Sunkavalli, William T. Freeman, Hao Tan

ICLR, 2026

project page / paper / code

Large-chunk test-time training scales nonlinear memory across language, 3D, and video.

RayZer: A Self-Supervised Large View Synthesis Model

Hanwen Jiang, Hao Tan, Peng Wang, Haian Jin, Yue Zhao, Sai Bi, Kai Zhang, Fujun Luan, Kalyan Sunkavalli, Qixing Huang, Georgios Pavlakos

ICCV, 2025

project page / paper / code

Self-supervised large-scale view synthesis from unposed image collections.

RigAnything: Template-Free Autoregressive Rigging for Diverse 3D Assets

Isabella Liu, Zhan Xu, Yifan Wang, Hao Tan, Zexiang Xu, Xiaolong Wang, Hao Su, Zifan Shi

ACM SIGGRAPH, 2025

project page / paper / code

Template-free autoregressive rigging for diverse 3D assets.

LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias

Haian Jin, Hanwen Jiang, Hao Tan, Kai Zhang, Sai Bi, Tianyuan Zhang, Fujun Luan, Noah Snavely, Zexiang Xu

ICLR, 2025 (Oral)

project page / paper / code

Feed-forward view synthesis with minimal hand-designed 3D inductive bias.

Long-LRM: Long-Sequence Large Reconstruction Model for Wide-Coverage Gaussian Splats

Ziwen Chen, Hao Tan, Kai Zhang, Sai Bi, Fujun Luan, Yicong Hong, Fuxin Li, Zexiang Xu

ICCV, 2025 (Highlight)

project page / paper / code

Wide-coverage scene reconstruction from 32 images in one feed-forward pass.

GS-LRM: Large Reconstruction Model for 3D Gaussian Splatting

Kai Zhang^*, Sai Bi^*, Hao Tan^*, Yuanbo Xiangli, Nanxuan Zhao, Kalyan Sunkavalli, Zexiang Xu

ECCV, 2024

project page / paper

Predicts object- and scene-level 3D Gaussians from sparse posed views.

MeshLRM: Large Reconstruction Model for High-Quality Mesh

Xinyue Wei, Kai Zhang, Sai Bi, Hao Tan, Fujun Luan, Valentin Deschaintre, Kalyan Sunkavalli, Hao Su, Zexiang Xu

arXiv, 2024

project page / paper

A feed-forward large reconstruction model for high-quality textured meshes.

PF-LRM: Pose-Free Large Reconstruction Model for Joint Pose and Shape Prediction

Peng Wang, Hao Tan, Sai Bi, Yinghao Xu, Fujun Luan, Kalyan Sunkavalli, Wenping Wang, Zexiang Xu, Kai Zhang

ICLR, 2024 (Spotlight)

project page / paper

Jointly estimates camera poses and object shape from sparse unposed images.

DMV3D: Denoising Multi-View Diffusion Using 3D Large Reconstruction Model

Yinghao Xu, Hao Tan, Fujun Luan, Sai Bi, Peng Wang, Jiahao Li, Zifan Shi, Kalyan Sunkavalli, Gordon Wetzstein, Zexiang Xu, Kai Zhang

ICLR, 2024

project page / paper / code

A 3D large reconstruction model inside a multi-view diffusion denoiser.

Instant3D: Fast Text-to-3D with Sparse-View Generation and Large Reconstruction Model

Jiahao Li, Hao Tan, Kai Zhang, Zexiang Xu, Fujun Luan, Yinghao Xu, Yicong Hong, Kalyan Sunkavalli, Greg Shakhnarovich, Sai Bi

ICLR, 2024

project page / paper

Fast text-to-3D via sparse-view diffusion and a large reconstruction model.

LRM: Large Reconstruction Model for Single Image to 3D

Yicong Hong, Kai Zhang, Jiuxiang Gu, Sai Bi, Yang Zhou, Difan Liu, Feng Liu, Kalyan Sunkavalli, Trung Bui, Hao Tan

ICLR, 2024 (Oral)

project page / paper

Introduced a scalable single-image-to-3D reconstruction model trained on massive data.

Scaling Data Generation in Vision-and-Language Navigation

Zun Wang, Jialu Li, Yicong Hong, Yi Wang, Qi Wu, Mohit Bansal, Stephen Gould, Hao Tan, Yu Qiao

ICCV, 2023 (Oral)

project page / paper / code

Scales vision-and-language navigation with 4.9 million instruction-trajectory pairs across 1,200+ environments.

How Much Can CLIP Benefit Vision-and-Language Tasks?

Sheng Shen^*, Liunian Harold Li^*, Hao Tan, Mohit Bansal, Anna Rohrbach, Kai-Wei Chang, Zhewei Yao, Kurt Keutzer

ICLR, 2022

paper / code

A systematic study of CLIP visual features for downstream vision-and-language tasks.

Unifying Vision-and-Language Tasks via Text Generation

Jaemin Cho, Jie Lei, Hao Tan, Mohit Bansal

ICML, 2021

paper / code

Unifies diverse vision-and-language tasks as conditional text generation.

LXMERT: Learning Cross-Modality Encoder Representations from Transformers

Hao Tan, Mohit Bansal

EMNLP-IJCNLP, 2019

paper / code

A cross-modal Transformer pretrained for vision-and-language reasoning.