Xindi(Cindy) Wu

I am Xindi Wu, a 4th year Ph.D. student in Computer Science at Princeton University, advised by Prof. Olga Russakovsky. Previously I was in Master of Science in Computer Vision at Carnegie Mellon University. I was very fortunate to be advised by Prof. Deva Ramanan and work with Prof. Min Xu and Prof. Haohan Wang and Prof. Kris Kitani. I was admitted to the Honors Youth Program at Xi'an Jiaotong University.

My research focuses on data for efficient multimodal machine learning, developing data-centric approaches that trace model performance back to training data composition and introduce data-driven capabilities to build more scalable and capable vision-language systems.

My work sits at the intersection of three core pillars:

I am happy to collaborate and answer questions about my research, feel free to send me an email. I especially encourage students from underrepresented groups to reach out.

Email /  CV /  Google Scholar /  LinkedIn /  GitHub  /  X

profile photo

News
2025
10/2025: Gave a poster presentation on Where is Motion From? Scalable Motion Attribution for Video Generation Models at ICCV Workshop on Reliable and Interactable World Models 2025 (Poster).
06/2025: Gave a talk at Datology AI Summer of Data Seminar on From Data to Capability: Data for efficient multimodal machine learning (Slides).
06/2025: Co-organizing the ICCV 2025 Workshop on Curated Data for Efficient Learning. We welcome submissions on data-centric approaches including data pruning, synthetic data, dataset distillation and more!
05/2025: Started my internship at Nvidia Spatial Intelligence Lab this summer in Santa Clara! ☀️
04/2025: Led a discussion on Compositionality in Vision-Language Discussion Group (Slides).
04/2025: Released COMPACT: COMPositional Atomic-to-Complex Visual Capability Tuning (Website).
02/2025: Attended NYC Vision Day and gave a poster presentation (Poster) with William Yang.
01/2025: Led a discussion on Data Influence in VisualAI Lab Reading Group (Slides).
2024
12/2024: Released ICONS: Influence Consensus for Vision-Language Data Selection (Website).
12/2024: Attended NeurIPS 2024 and gave a poster presentation on ConceptMix (Poster).
10/2024: Our paper Corgi: Cached Memory Guided Video Generation is accepted to WACV 2025.
09/2024: Attended ECCV 2024 and gave a spotlight talk on ConceptMix (Slides) at Knowledge in Generative Models Workshop; and our Vision-Language Dataset Distillation received best paper award at Dataset Distillation Workshop.
09/2024: Our paper ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty is accepted to NeurIPS D&B 2024.
07/2024: Our paper Vision-Language Dataset Distillation is accepted to TMLR 2024.
07/2024: Gave a talk on Compositional Generation Evaluation (Slides) at Google Research.
06/2024: Attended CVPR 2024 and gave a talk on Scaling Down before Scaling Up: Recent Progress on Dataset Distillation (Slides | Video) at CVPR 2024 Dataset Distillation Workshop.
02/2024: Led a discussion on Scaling Law in the PLI Vision-Language Reading Group (Slides).
02/2024: I am TAing for COS 429 Computer Vision in Spring 2024.
01/2024: Passed my general exam (quals)! (Video | Slides | Reading List). Huge thanks to my committee members Olga Russakovsky, Szymon Rusinkiewicz and Adji Bousso Dieng for their support and feedback!
2023
09/2023: I am TAing for COS 597O Advanced Topics in Computer Science: Deep Generative Models, with focus on Methods, Applications & Societal Considerations in Fall 2023.
06/2023: Attended CVPR 2023 and presented Pix2Map (Poster).
05/2023: Started my internship at Meta Reality Lab in Redmond, WA on multimodal generative models.
02/2023: One paper accepted to CVPR 2023: Pix2Map: Cross-modal Retrieval for Inferring Street Maps from Images.
2022
08/2022: Started my PhD journey at Princeton University! 💪
Selected Publications & Preprints
Where is Motion From? Scalable Motion Attribution for Video Generation Models.
Xindi Wu, Despoina Paschalidou, Jun Gao, Antonio Torralba, Laura Leal-Taixé, Olga Russakovsky, Sanja Fidler, Jonathan Lorraine
Poster | Arxiv (Coming soon)
ICCV Workshop on Reliable and Interactable World Models 2025
Beyond Objects: Contextual Synthetic Data Generation for Fine-Grained Classification.
William Yang, Xindi Wu, Zhiwei Deng, Esin Tureci, Olga Russakovsky
Arxiv | Code
Preprint 2025
Actions as Language: Fine-Tuning VLMs into VLAs Without Catastrophic Forgetting.
Asher J. Hancock, Xindi Wu, Lihan Zha, Olga Russakovsky, Anirudha Majumdar
Website | Arxiv
Preprint 2025
COMPACT: COMPositional Atomic-to-Complex Visual Capability Tuning.
Xindi Wu*, Hee Seung Hwang*, Polina Kirichenko, Olga Russakovsky
Website | Arxiv | Code | Thread
Preprint 2025
ICONS: Influence Consensus for Vision-Language Data Selection.
Xindi Wu, Mengzhou Xia, Rulin Shao, Zhiwei Deng, Pang Wei Koh and Olga Russakovsky
Website | Arxiv | Code | Huggingface | Thread.
Preprint 2024
ConceptMix: A Compositional Image Generation Benchmark with Controllable Difficulty.
Xindi Wu*, Dingli Yu*, Yangsibo Huang*, Olga Russakovsky and Sanjeev Arora
Website | Arxiv | Slides | Poster | Code | Thread.
NeurIPS Datasets & Benchmarks 2024, ECCV Knowledge in Generative Models Workshop (Spotlight)
Corgi: Cached Memory Guided Video Generation.
Xindi Wu, Uriel Singer, Zhaojiang Lin, Andrea Madotto, Xide Xia, Yifan Xu, Paul Crook, Xin Luna Dong and Seungwhan Moon.
Paper | Supp | IEEE Paper Site | Poster | Bibtex
WACV 2025
Vision-Language Dataset Distillation.
Xindi Wu, Byron Zhang, Zhiwei Deng and Olga Russakovsky.
Website | Arxiv | TMLR Paper Site | Video (45 mins) | Slides(canva, pdf) | Poster | Code | Thread.
TMLR 2024, ECCV Dataset Distillation Workshop (Best Paper)
Pix2Map: Cross-modal Retrieval for Inferring Street Maps from Images.
Xindi Wu, KwunFung Lau, Francesco Ferroni, Aljoša Ošep and Deva Ramanan
Website | Arxiv | CVPR Paper Site | Youtube Video | Poster | Slides | Thread
CVPR 2023
Toward Learning Robust and Invariant Representations with Alignment Regularization and Data Augmentation.
Haohan Wang, Zeyi Huang, Xindi Wu and Eric P. Xing.
KDD 2022
Ego4D: Around the World in 3,000 Hours of Egocentric Video.
Kristen Grauman, et al .
CVPR 2022
High Frequency Component Helps Explain the Generalization of Convolutional Neural Networks.
Haohan Wang, Xindi Wu, Zeyi Huang, and Eric P. Xing.
Arxiv | CVPR Paper Site | Youtube Video | Slides | Code.
CVPR 2020 (Oral)
Transferable Adversarial Attacks on Deep Reinforcement Learning with Domain Randomization.
Xinlei Pan, Yulong Cao, Xindi Wu, Eric Zelikman, Yanan Sui, Rudrasis Chakraborty, and Ronald S. Fearing .
CVPR Workshop, 2020
Regularized Adversarial Training (RAT) for Robust Cellular Electron Cryo Tomograms Classification
Xindi Wu, Yijun Mao, Haohan Wang, Xiangrui Zeng, Xin Gao, Eric P. Xing and Min Xu.
BIBM 2019 (Oral)
Template-Based and Template-Free Approaches in Cellular Cryo-Electron Tomography Structural Pattern Mining.
Xindi Wu, Xiangrui Zeng, Zhenxi Zhu, Xin Gao, and Min Xu
Computational Biology, Codon Publications, Brisbane, Australia, 2019
Deep Self-Paced Learning for Semi-Supervised Person Re-Identification Using Multi-View Self-Paced Clustering
Xiaomeng Xin, Xindi Wu, Yuechen Wang, and Jinjun Wang.
ICIP 2019
Multitask Learning With Enhanced Modules
Zishuo Zheng, Yadong Wei, Zixu Zhao, Xindi Wu, Zhengcheng Li, and Pengju Ren.
DSP 2018
Education
Princeton University, USA
Ph.D. in Computer Science • Aug. 2022 to Now
Advisor: Prof. Olga Russakovsky
Carnegie Mellon University, USA
M.S. in Computer Vision • Sept. 2020 to Dec. 2021
Advisor: Prof. Deva Ramanan
Xi'an Jiaotong University, China
B.Sc. in Computer Science • Sep. 2016 to July 2020
Advisor: Prof. Jinjun Wang, Prof. Pengju Ren
National University of Singapore, Singapore
Summer session • Human Computer Interaction •July 2017 to Aug. 2017

Experiences
Spatial Intelligence Lab, NVIDIA
Research Scientist Intern • May 2025 - Now
Mentors & Advisors: Dr. Jonathan Lorraine, Dr. Despoina Paschalidou, Dr. Jun Gao, Prof. Laura Leal-Taixé, Prof. Antonio Torralba, Prof. Sanja Fidler
Meta Reality Labs, Meta.
Research Scientist Intern • May. 2023 to Aug. 2023
Mentors: Dr. Shane Moon, Dr. Uriel Singer Dr. Luna Dong, Dr. Zhaojiang Lin, Dr. Paul Crook, Dr. Andrea Madotto Dr. Ethan Xu
Perception ML Team, Snap Inc.
Machine Learning Engineer • Feb. 2022 to July 2022
Research Intern • May. 2021 to Aug. 2021
Mentors: Dr. Alireza Zareian and Dr. Chen Wang
CMU Argo AI Center for Autonomous Vehicle Research, ArgoAI
Capstone Project • Feb. 2021 to Dec 2021
Advisors: Prof. Deva Ramanan. Mentors: Dr. Aljoša Ošep and Dr. Francesco Ferroni
Megvii (Face ++) Research, Megvii
Computer Vision Research Intern • June 2020 to Sept. 2020
Mentor: Banghuai Li
Computational Biology Department, Carnegie Mellon University, USA
Research Intern • March 2019 to Feb. 2020
Advisor: Prof. Min Xu
Institute of Artificial Intelligence and Robotics, Xi'an Jiaotong University, China
Research Intern • Dec. 2017 to Feb. 2019
Advisors: Prof. Pengju Ren and Prof. Jinjun Wang

Teaching
TA: COS 429 Computer Vision, Spring 2024, Princeton, with instructor: Vikram V. Ramaswamy and Felix Heide
TA: COS 597O Advanced Topics in Computer Science: Deep Generative Models: Methods, Applications & Societal Considerations. Fall 2023, Princeton, with instructor: Adji Bousso Dieng.
Professional Activities & Outreach

Reviewer: NeurIPS 25'/24'/23', ICLR 25'/24', ICML 25'/24', CVPR 25'/24'/23'/22', ICCV 25'/23', ECCV 24'/22', TMLR, ICRA 24', ACCV 24', ICLR 23' Workshop ME-FoMo, NeurIPS Interpolate Workshop 22', BMVC 20', IJCAI 20'.
Workshop Co-organizer: ICCV 2025 Workshop on Curated Data for Efficient Learning.
Committee Member: Diversity, Equity and Inclusion Committee at Robotics Institute, CMU.

Website source from Jon Barron