<div style="display: inline-block; width: 100%;"> <div style="float:left; width: 80%;"> ## Ziteng Gao I am a last-year PhD student at the National University of Singapore (NUS), currently working at the [Show Lab](https://sites.google.com/view/showlab/home) under the supervision of [Prof. Mike Z. Shou](https://cde.nus.edu.sg/ece/staff/shou-zheng-mike/). My current research interests focus on __vision generation__. Previously, I received my M.Sc. degree from Nanjing University in 2022, advised by [Prof. Limin Wang](https://wanglimin.github.io/). I obtained my B.Sc. degree from the same institution in 2019 as a member of the Computer Science Elite Class. `gzt` `[at]` `outlook.com` [[github]](https://github.com/sebgao) [[google scholar]](https://scholar.google.com/citations?user=fbSH2CgAAAAJ&hl=en) <span style="color: red;">I am always seeking full-time research or engineering positions in industry starting late 2026. Feel free to reach out via the email above if you think I would be a good fit for your team.</span> </div> <div style="float:right; width: 20%;">  </div> </div> ## # ## Selected Papers & Preprints, [Full List](https://scholar.google.com/citations?user=fbSH2CgAAAAJ&hl=en) #### D-AR: Diffusion via Autoregressive Models _ __Ziteng Gao__, Mike Zheng Shou_ ICLR, 2026 <span style="color: red;">[new]</span> [[arxiv]](https://arxiv.org/abs/2505.23660) [[code]](https://github.com/showlab/d-ar) <br> #### Pixnerd: Pixel Neural Field Diffusion _ Shuai Wang, Ziteng Gao, Chenhui Zhu, Weilin Huang, Limin Wang_ ICLR, 2025 <span style="color: red;">[new]</span> [[arxiv]](https://arxiv.org/abs/2507.23268) [[code]](https://github.com/MCG-NJU/PixNerd) <br> #### Sparse Image Synthesis via Joint Latent and RoI Flow _ __Ziteng Gao__, Jay Zhangjie Wu, Mike Zheng Shou_ NeurIPS, 2025 <span style="color: red;">[new]</span> [paper & code to be released] <br> <br> #### Bootstrapping SparseFormers from Vision Foundation Models _ __Ziteng Gao__, Zhan Tong, Kevin Qinghong Lin, Joya Chen, Mike Zheng Shou_ CVPR, 2024 [[arxiv]](https://arxiv.org/abs/2312.01987) [[pdf]](https://arxiv.org/pdf/2312.01987.pdf) [[code]](https://github.com/showlab/sparseformer) <br> #### SparseFormer: Sparse Visual Recognition via Limited Latent Tokens _ __Ziteng Gao__, Zhan Tong, Limin Wang, Mike Zheng Shou_ ICLR, 2024 [[arxiv]](https://arxiv.org/abs/2304.03768) [[pdf]](https://arxiv.org/pdf/2304.03768.pdf) [[code]](https://github.com/showlab/sparseformer) <br> #### AdaMixer: A Fast-Converging Query-Based Object Detector _ __Ziteng Gao__, Limin Wang, Bing Han, Sheng Guo_ CVPR, 2022 (__oral__) [[arxiv]](https://arxiv.org/abs/2203.16507) [[pdf]](https://arxiv.org/pdf/2203.16507.pdf) [[code]](https://github.com/MCG-NJU/AdaMixer) [[slides]](https://github.com/MCG-NJU/AdaMixer/blob/main/adamixer_cvpr_2022_keynote.pdf) <br> #### Mutual Supervision for Dense Object Detection _ __Ziteng Gao__, Limin Wang, Gangshan Wu_ ICCV, 2021 [[arxiv]](https://arxiv.org/abs/2109.05986) [[pdf]](https://openaccess.thecvf.com/content/ICCV2021/papers/Gao_Mutual_Supervision_for_Dense_Object_Detection_ICCV_2021_paper.pdf) <br> #### Boundary-Aware Cascade Networks for Temporal Action Segmentation _Zhenzhi Wang, Ziteng Gao, Limin Wang, Zhifeng Li, Gangshan Wu_ ECCV, 2020 [[pdf]](http://www.ecva.net/papers/eccv_2020/papers_ECCV/papers/123700035.pdf) [[code]](https://github.com/MCG-NJU/BCN) <!-- [[bibtex]](./bibtex/Gao_2019_ICCV.bibtex) --> <br> #### LIP: Local Importance-based Pooling _ __Ziteng Gao__, Limin Wang, Gangshan Wu_ ICCV, 2019 <!-- ##### TL;DR: a learnable, simple, generalized but effective pooling method --> [[arxiv]](https://arxiv.org/abs/1908.04156) [[pdf]](http://openaccess.thecvf.com/content_ICCV_2019/papers/Gao_LIP_Local_Importance-Based_Pooling_ICCV_2019_paper.pdf) [[code]](https://github.com/sebgao/LIP) <!-- [[bibtex]](./bibtex/Gao_2019_ICCV.bibtex) --> # <!-- ### Personal Project [cTensor](https://github.com/sebgao/cTensor) -- a super lightweight deep learning library based on NumPy in PyTorch fashion, which supports convolutional operators. ### Friend [Hanqing Wu](https://hanqing-wu.weebly.com) -->