CosmoVAE: Variational Autoencoder for CMB Image Inpainting
Yi K, Guo Y, Fan Y, et al. Cosmo vae: Variational autoencoder for cmb image inpainting[C]//2020 International Joint Conference on Neural Networks (IJCNN). IEEE, 2020: 1-7.
CV
Education
- Ph.D in Data Visualization, Tongji University, 2020-2024
- Intelligent Big Data Visualization Lab, Advisor: Prof. Nan Cao
- Research Directions: Language Models, Natural Language Processing, Data Visualization, Human-Computer Interaction
- Outstanding Graduate
- M.S. in Financial Mathematics, University of New South Wales, 2018-2019
- School of Mathematics and Statistics, Advisor: Associate Prof. Yuguang Wang
- Research Directions: Bayesian Analysis, Graph Neural Networks
- B.S. in Mathematics and Finance, University of Toronto, 2013-2017
- Dual Degree in Mathematics and Finance
Work experience
- Feb 2025 - Present: Senior Algorithm Scientist
- Global Analysis and Data Science (GADS), PayPal
- Focus on Agentic AI technologies for risk control and SDLC business scenarios
- Developed ChatBI analytics system and QE testing system using multi-agent architecture
- Responsible for agent algorithm frameworks, RAG systems, and data visualization
- Aug 2024 - Dec 2024: Algorithm Research Intern
- ByteDance, Doubao Large Speech Model Team
- Participated in designing distributed machine learning platform for speech/music algorithms
- Built large-scale data processing pipelines and developed multimodal data processing models
- Jan 2020 - Dec 2024: PhD Researcher
- Intelligent Big Data Visualization Lab, Tongji University
- Research on structured data question-answering and automatic generation techniques
- Focus on large language models for structured data tasks and narrative visualization
- Served as reviewer for top conferences: VIS 2021-2023, CHI 2022-2023, AAAI 2022
- Aug 2020 - Dec 2020: Quantitative Research Intern
- China Merchants Fund, Mentor: Dr. Wei Xiang
- Developed AI-driven investment system with visualization components
- Implemented reinforcement learning-based investment algorithms
- Summer 2017: Intern Industry Analyst
- Shenwan Hongyuan Securities Research
- Mentor: Jing Liu
- Fall 2019: Intern Algorithm Engineer
- GDS Services Ltd.
- Mentor: Yu Chen
Publications
Talk2Data: A Natural Language Interface for Exploratory Visual Analysis via Question Decomposition
Guo Y, Shi D, Guo M, et al. Talk2Data: A Natural Language Interface for Exploratory Visual Analysis via Question Decomposition[J]. arXiv e-prints, 2021: arXiv: 2107.14420.
Survey on visual analysis of event sequence data
Guo Y, Guo S, Jin Z, et al. Survey on visual analysis of event sequence data[J]. IEEE Transactions on Visualization and Computer Graphics, 2021, 28(12): 5091-5112.
Datamator: An Intelligent Authoring Tool for Creating Datamations via Data Query Decomposition
Guo Y, Cao N, Cai L, et al. Datamator: An Authoring Tool for Creating Datamations via Data Query Decomposition[J]. Applied Sciences, 2023, 13(17): 9709.
Chart2Vec: A Universal Embedding of Context-Aware Visualizations
Chen Q, Chen Y, Shuai W, et al. Chart2Vec: A Universal Embedding of Context-Aware Visualizations[J]. arXiv preprint arXiv:2306.08304, 2023.
Urania: Visualizing Data Analysis Pipelines for Natural Language-Based Data Exploration
Guo Y, Cao N, Qi X, et al. Urania: Visualizing Data Analysis Pipelines for Natural Language-Based Data Exploration[J]. arXiv preprint arXiv:2306.07760, 2023.
Talks
Visual Analytics Sytem
Talk at Tongji University iDVX × HKUST Doctoral Forum, Shanghai, CHN
Survey on visual analysis of event sequence data
Conference proceedings talk at VIS2022, Oklahoma City, USA
Enhancing Natural Language-Based Data Exploration with Analysis Pipeline Illustration
Poster at VIS2023, Melbourne, Australia
Open Source Projects
- Narrative Chart (Jun 2022 - Present): Open-source JavaScript library for creating visual charts to facilitate data storytelling
- Designed and developed from scratch, officially released in June 2022
- Responsible for animation module and Unit Visualization components
- Available at: https://narchart.github.io/
- Calliope (Oct 2020 - Present): Intelligent Data Visualization Story Generation Platform
- Online platform for generating, creating, editing, and sharing data visualization content
- Integrated GPT-3.5 into story generation algorithm
- Available at: https://datacalliope.com/
Presentations
- Visual Analytics Systems, Tongji University IDvX Lab & Hong Kong University of Science and Technology PhD Student Forum, June 2021
- Survey on Visual Analysis of Event Sequence Data, VIS 2022 Conference, October 2022
- Natural Language-Oriented Data Visualization, Tongji University IDvX Lab Second Open Day, August 2023
- Enhancing Natural Language-Based Data Exploration with Analysis Pipeline Illustration, VIS 2023 Conference, October 2023
Technical Skills
- Algorithm Research: Proficient in deep learning frameworks (PyTorch, Hugging Face, Langgraph, CrewAI)
- Backend Development: Proficient in Flask framework, RESTful APIs development
- Front-End Development: Proficient in D3.js, familiar with React.js
- Programming Languages: Proficient in Python, familiar with JavaScript, SQL
- Project Operations: Experience in Linux server administration, Docker, VM virtualization
Service and leadership
- Reviewer of VIS 2021-2023, CHI 2022-2023, AAAI 2022, IEEE Transactions on Big Data
- Reviewer of ChinaVIS 2023