Xichen Pan

Office: 424, 60 5th Ave, New York, NY 10011

Work Email: xichenpan [at] nyu [dot] edu

Personal Email: xcpan [dot] mail [at] gmail [dot] com

Bio

I am a first-year Ph.D. student of Computer Science at NYU Courant, advised by Prof. Saining Xie. I interned at Microsoft Research Asia (2022-2023 with Dr. Li Dong), Alibaba Group (2022 with Dr. Pengda Qin), and Horizon Robotics (2021-2022 with Yichen Gong). Previously, I obtained my bachelor’s degree in Computer Science from Shanghai Jiao Tong University (SJTU) and won the Best Thesis Award. I was fortunately advised by and maintain a close connection with Prof. Zhouhan Lin at SJTU.

Research Interest

Generative Models

Designing more controlable and high fidelity methods for image, video, and 3D generation, with a focus on:

Preserving spatial and temporal consistency
Leveraging text-to-image priors for advanced applications

Multimodal Learning

Developing vision-language models for vision-centric applications, focusing on representation learning and self-supervised pre-training

News

202420232022

[02/2024] Our paper was accepted by CVPR 2024, check it out here.

[02/2024] I will join Meta GenAI as a Research Scientist Intern in 2024 Summer. See you in Menlo Park!

[01/2024] Our paper was accepted by ICLR 2024, check it out here. See you in Vienna!

[10/2023] Our paper was accepted by WACV 2024 as Oral, check it out here.

[09/2023] Excited to start my CS Ph.D. at NYU Courant advised by Prof. Saining Xie.

[12/2022] Glad to work with Dr. Li Dong and Dr. Furu Wei at Microsoft Research Asia for the upcoming year, leading up to Fall 2023.

[06/2022] My bachelor thesis won Best Thesis Award in SJTU! Thanks my advisor Prof. Zhouhan Lin, checkout the honor roll.

[02/2022] Our paper was accepted by ACL 2022 Main Conference, check out full paper.

Education

New York University Courant Institute

Sept. 2023 -- Present

Ph.D. in Computer Science, advised by Prof. Saining Xie

Shanghai Jiao Tong University

Sept. 2018 -- June 2022

B.Eng. in Computer Science (Outstanding Graduate of Class 2022), advised by Prof. Zhouhan Lin

Publications & Manuscripts

Image Sculpting: Precise Object Editing with 3D Geometry Control

Jiraphon Yenphraphai, Xichen Pan, Sainan Liu, Daniele Panozzo, Saining Xie

CVPR 2024 arXiv Code Project Page

Kosmos-G: Generating Images in Context with Multimodal Large Language Models

Xichen Pan, Li Dong, Shaohan Huang, Zhiliang Peng, Wenhu Chen, Furu Wei

ICLR 2024 arXiv Code Project Page

Synthesizing Coherent Story with Auto-Regressive Latent Diffusion Models

Xichen Pan, Pengda Qin, Yuhong Li, Hui Xue, Wenhu Chen

WACV 2024 (Oral, Top 6% of accepted papers) arXiv Code

Multimodal Audio-Visual Speech Recognition System Based On Pre-trained Models

Xichen Pan

Bachelor thesis at Shanghai Jiao Tong University (Best Thesis Award, 1st/150) News Honor Roll

Leveraging Unimodal Self-Supervised Learning for Multimodal Audio-visual Speech Recognition

Xichen Pan, Peiyu Chen, Yichen Gong, Helong Zhou, Xinbing Wang, Zhouhan Lin

ACL 2022 Main Conference arXiv Code

Experience

Microsoft Research Asia Image title

Dec. 2022 – Sept. 2023

StarBridge Program Research Assistant

Alibaba Group Image title

Sept. – Dec. 2022

Research Intern

Horizon Robotics Image title

Apr. 2021 – July 2022

Research Intern

John Hopcroft Center for Computer Science, Shanghai Jiao Tong University

Apr. 2021 – June 2022

Research Intern

NSF Center for Big Learning, University of Florida Image title

July – Sept. 2020

Research Intern

Selected Projects

Open CS Application

An open-source GitHub page built for reference in selecting CS programs in north America. The page is powered by Material for MkDocs and supports collaboration through Pull Requests and GitHub Actions.