CV
Updated September 12, 2026
Research Interests
Computational Social Science, Network Science and Statistical Network Models, LLM-Based Agentic Systems for Scientific Discovery, Reliability and Grounding of AI-Generated Research
Education
University of Southern California
Los Angeles, CA, USA
- Ph.D. Candidate, Industrial and Systems Engineering (Aug. 2023 - Expected May 2027)
- M.S., Analytics (Aug. 2021 - Dec. 2022)
- Dissertation: Discovery of Network Formation Processes via LLM-Guided Exponential Random Graph Models
- Advisor: Mayank Kejriwal
- Key Courses: Design of Experiments and Quality Engineering, Stochastic Processes, Decision Analysis, Reinforcement Learning, Linear Optimization, Stochastic Optimization, Data Management, Data Mining, Text Analytics
The Chinese University of Hong Kong, Shenzhen
Shenzhen, China
- B.S., Statistics (Sep. 2016 - May 2020)
- Key Courses: Probability and Statistics, Regression Analysis, Statistical Inference, Survey Sampling, Nonparametric Statistics, Categorical Data Analysis, Statistical Modelling in Financial Markets, Stochastic Calculus
Honors and Awards
- Jenny Wang Excellence in Teaching Award, USC Viterbi School of Engineering (2024)
- Outstanding Teaching Assistant of the Year, USC Daniel J. Epstein Department of Industrial and Systems Engineering (2024)
Publications
- Kejriwal, M., & Sun, Y. (2026). Building discoverative AI ecosystems for supporting high-quality agentic science. AIAS+ 2026. Forthcoming.
- Sun, Y., & Kejriwal, M. (2026). Modeling complex associations between socio-demographic factors and climate policies using Conditional Inference Trees. Humanities and Social Sciences Communications, 13.
- Sun, Y., & Kejriwal, M. (2026). Structural modeling of campaign finance decisions in the U.S. House of Representatives. PLOS Complex Systems, 3(5), e0000104.
- Sun, Y., He, D., & Kejriwal, M. (2026). Examining persistent inequities in financial complaint resolution for older Americans and veterans in the United States. PNAS Nexus, 5(4), pgag086.
- Sun, Y., Shen, K., & Kejriwal, M. (2025). Backlash or reinforcement? Donald Trump’s 2017 inauguration and shifting climate beliefs in the United States. npj Climate Action, 4, Article 26.
- Sun, Y. (2025). Scalable estimation of exponential random graph models on large networks. Companion Proceedings of the ACM Web Conference 2025, 721-724.
- Gawin, C., Sun, Y., & Kejriwal, M. (2025). Navigating semantic relations: Challenges for language models in abstract common-sense reasoning. Companion Proceedings of the ACM Web Conference 2025, 971-975.
- Sun, Y., & Kejriwal, M. (2024). A study of firm-switching of inventors in Big Tech using public patent data. In Social Network Analysis and Mining Applications in Healthcare and Anomaly Detection. Springer.
- Sun, Y., & Kejriwal, M. (2023). A structural study of Big Tech firm-switching of inventors in the post-recession era. Proceedings of the International Conference on Advances in Social Networks Analysis and Mining, 670-677.
- Sun, Y., & Kejriwal, M. (2023). DeepGraph: Multi-cluster interactive visualization of complex networks in a learned representation space. Proceedings of the International Conference on Advances in Social Networks Analysis and Mining, 427-430.
Under Review
- Patil, V., Bacha, S. V., Yamani, R., Sun, Y., & Kejriwal, M. (2026). Beyond the star rating: LLM-assisted aspect analysis and large-scale text classification of 4.7 million restaurant reviews. Revised manuscript under review at PLOS ONE.
- Sun, Y., & Kejriwal, M. (2026). Online sex-work advertising on Backpage in Southeast Asia, 2013-2018: A retrospective analysis of posting, place, and price. Revision under review at SN Social Sciences.
- Sun, Y., & Kejriwal, M. (2026). FORGE: An agentic system for ERGM specification and interpretation in social networks. Submitted to WSDM 2027.
- Sun, Y., et al. (2026). Computational social science for harm reduction: Privacy-preserving analysis of market structure in counter-trafficking work. Submitted to the AAAI 2027 Artificial Intelligence for Social Impact Track.
Under Preparation
- Sun, Y., & Kejriwal, M. (2026). Hallucinated references in AI-generated publications: A cross-corpus audit. Manuscript in preparation.
- Sun, Y., & Kejriwal, M. (2026). Interactive exploration of tie-formation hypotheses in complex networks. Manuscript in preparation (demonstration paper).
- Sun, Y., & Kejriwal, M. (2026). Coordination and fragmentation in Latin American and Caribbean online sex work markets. Manuscript in preparation.
- Sun, Y., & Kejriwal, M. (2026). A framework for organized control at the account level in online sex trafficking detection. Manuscript in preparation.
Research Funding
- DARPA MAGICS, $300,000. Primary proposal author; PI: Mayank Kejriwal (2026)
- Kroner Family Foundation. Philanthropic gift supporting research on human trafficking (2024-2026)
Selected Presentations
- Poster Presentation: “Building reliable PhD-level agents for computational research,” Agentic AI Summit, University of California, Berkeley, Berkeley, CA, Aug. 1-2, 2026.
- Poster Presentation: “Mapping Collaboration in AI-Driven Firms: A Co-Inventor Dataset (2020-2025),” International Conference of the Computational Social Science Society of the Americas, Washington, D.C., Nov. 6-9, 2025.
- Poster Presentation: “Reviewer Rating Variability, Reviewer Confidence, and Language Model-Based Sentiment Prediction of Machine Learning Conference Submissions,” 10th International Congress on Peer Review and Scientific Publication, Chicago, IL, Sep. 3-5, 2025.
- Oral Presentation: “Collaboration Diversity and Paper Success: A Network Analysis of Co-authorship in a Major Machine Learning Conference,” INSNA Sunbelt Conference, Paris, France, Jun. 23-29, 2025.
- Poster Presentation: “Scalable estimation of exponential random graph models on large networks,” The Web Conference, May 13, 2025.
- Oral Presentation: “Navigating semantic relations: Challenges for large language models in abstract common-sense reasoning,” The Web Conference, May 13, 2025.
- Oral Presentation: “AI Reveals Systemic Financial Inequities: A Study of CFPB Complaints for Older Adults and Veterans,” USC ShowCAIS Symposium, Apr. 2025.
- Oral Presentation: “Understanding the drivers of human trafficking using massive multi-partite network modeling and analytics,” USC Cloudwalkers Premiere, Mar. 2025.
- Oral Presentation: “A structural study of Big Tech firm-switching of inventors in the post-recession era,” Advances in Social Networks Analysis and Mining, Nov. 7, 2023.
- Oral Presentation: “DeepGraph: Multi-cluster interactive visualization of complex networks in a learned representation space,” Advances in Social Networks Analysis and Mining, Nov. 6, 2023.
- Poster Presentation: “Modelling campaign finance data using network science to map structures and flows of influence,” Southern California Applied Mathematics Symposium, Apr. 22, 2023.
Media Coverage
Public Writing
- “5 lessons from 20 years of climate surveys: Political party matters, but it doesn’t explain everything”, The Conversation, Jul. 21, 2026. Syndicated by The Washington Post, NationofChange, and ArcaMax.
Research Coverage
- “A majority of Americans don’t actually fit either climate camp,” Anthropocene Magazine, Aug. 4, 2026.
- “Financial complaint delays hit seniors and veterans hardest, with gaps widening over time,” Phys.org, Apr. 16, 2026.
- “Older Americans and veterans wait longer for financial justice,” USC Information Sciences Institute, Apr. 16, 2026.
- “Big Tech inventor mobility leads to higher productivity”, USC Information Sciences Institute, Nov. 7, 2023.
Public Scholarship
Global Trafficking Intelligence Hub (2024-Present)
Founder and Lead Developer, globaltrafficking.org
Built and maintain a public knowledge hub on human trafficking, hosted by the USC Artificial Intelligence and Complex Systems group with support from the Kroner Family Foundation. The hub curates international reports, country- and region-level assessments, legal frameworks, datasets, and peer-reviewed research across eight world regions, with source-dated entries and a searchable research interface for practitioners, policymakers, and researchers.
Research Software and Data
- Campaign Finance Networks and Covariates, 2009-2022: Public data and code accompanying the PLOS Complex Systems article. github.com/yidans/Campaign-Finance
Teaching Experience
University of Southern California — Teaching Assistant
- ISE 547: Applied Generative Artificial Intelligence for Enterprises (Fall 2025)
- ISE 529: Predictive Analytics (Fall 2025)
- ISE 561: Economic Analysis of Engineering Projects (Summer 2025)
- ISE 540: Text Analytics (Fall 2023, Spring 2025, Summer 2025)
Student Mentoring
- Shama Maganur, Montana State University (Spring 2026)
- Richard Joungyoon Kim, CURVE Program, USC (Fall 2024-2025)
- Angel Otters, Directed Research, USC (Fall 2024)
- Daiqi He, Directed Research, USC (Summer 2024-2025)
- Erica Okeh, Viterbi SURE Program, Howard University (Summer 2024)
- Herry Wang, Directed Research, USC (Summer 2024)
- Shree Vaishnavi Bacha, Directed Research, USC (Spring 2024-2025)
- Vishal Patil, Directed Research, USC (Spring 2024-2025)
- Yining Ma, Directed Research, USC (Spring 2024)
- Revanth Yamani, Directed Research, USC (Spring 2024)
- John Lim, REU Summer Program, Northwestern University (Summer 2023)
Professional Service
- Conference Reviewer: Conference on Empirical Methods in Natural Language Processing (EMNLP) (2026)
- Journal Reviewer: Social Network Analysis and Mining, Frontiers in Political Science, and Frontiers in Sociology
Research Experience
GRAIL, Los Angeles, CA
- Research Scientist (May 2026 - Aug. 2026)
- Built LLM-based agentic systems for scientific research workflows, including an autonomous research pipeline that decomposes questions into executable steps and orchestrates tool calls across retrieval, analysis, and code execution; a retrieval-augmented grant recommender; and a hallucination checker that audits generated scientific text against source documents and bibliographic databases, with an accompanying evaluation harness and error taxonomy.
USC Information Sciences Institute, Marina del Rey, CA
- Research Engineer, Artificial Intelligence and Complex Systems Group (Jan. 2023 - Aug. 2023)
- Conducted applied research in the Artificial Intelligence and Complex Systems group; designed and ran controlled experiments, implemented model variants and evaluation pipelines, and contributed to academic publications through manuscript drafting and preparation of experimental results.
Datago Technology Limited, Hong Kong
- Quantitative Researcher Intern (May 2021 - Sep. 2021)
- Built data pipelines to extract and standardize text features for sentiment-based factors; implemented factor construction logic from raw forum data and ran signal diagnostics and walk-forward backtests.
Technical Skills
- Methods: Exponential Random Graph Models, Statistical Network Inference, Community Detection and Centrality Analysis, Graph Embeddings, Supervised and Unsupervised Learning, Text Classification and Topic Modeling
- LLM and Agentic Systems: Agent Orchestration, Retrieval-Augmented Generation, Tool Use and Function Calling, LLM Evaluation Pipelines
- Software: Python (PyTorch, Transformers, scikit-learn, pandas), R (ergm, statnet), SQL, Neo4j, Gephi