👋 I am a Ph.D. student in the Civil and Environmental Engineering Department at the University of California, Los Angeles (UCLA), where I focus on cutting-edge research in intelligent transportation systems and autonomous driving. I received my M.S. degree in Computer Science from New York University in 2024 and my B.E. degree in Computer Science and Engineering from the University of California, Davis.

My research interests span large language models, reinforcement learning, and deep learning, with particular emphasis on applications to smart mobility systems, autonomous driving, and intersection safety. I am passionate about developing AI-driven solutions that can make transportation safer and more efficient.

🔥 News

📝 Selected Publications

MobiVerse: Scaling Urban Mobility Simulation with Hybrid Lightweight Domain-Specific Generator and Large Language Models

Yifan Liu, Xishun Liao, Haoxuan Ma, Jonathan Liu, Rohan Jadhav, Jiaqi Ma

A novel hybrid framework that combines lightweight domain-specific generators with LLMs for efficient and scalable urban mobility simulation, demonstrated through a case study in Westwood, Los Angeles, successfully simulating 53,000 agents with dynamic adaptations to environmental changes.

IEEE ITSC, 2025 | 📄 Paper

Human Mobility Modeling with Household Coordination Activities under Limited Information via Retrieval-Augmented LLMs

Yifan Liu, Xishun Liao, Haoxuan Ma, Brian Yueshuai He, Chris Stanford, Jiaqi Ma

A retrieval-augmented LLM framework that generates realistic activity chains with household coordination using only publicly accessible statistical information, addressing key limitations in mobility modeling for data-scarce regions.

IEEE ITSC, 2025 | 📄 Paper

Socially-Aware Robot Navigation Enhanced by Bidirectional Natural Language Conversations Using Large Language Models

Congcong Wen, Yifan Liu, Geeta Chandra Raju Bethala, Shuaihang Yuan, Hao Huang, Yu Hao, Mengyu Wang, Yu-Shen Liu, Anthony Tzes, Yi Fang

A novel HSAC-LLM framework that integrates deep reinforcement learning with large language models to enable robots to proactively communicate with pedestrians through natural language for collision avoidance and socially-aware navigation.

IEEE IROS, 2025 | 📄 Paper

CDA.AI for OpenCDA: AI pathways for cooperative driving automation research

Xu Han, Zhaoliang Zheng, Zewei Zhou, Yun Zhang, Tiahui Cai, Yifan Liu, Hao Xiang, Camila Correa-Jullian, Zonglin Meng, Zhiyu Huang, Letian Gao, Xin Xia, Jiaqi Ma

A comprehensive AI framework for Cooperative Driving Automation that integrates connectivity, vision-language models, and multi-agent systems to enable scalable and interpretable autonomous driving solutions through advanced perception, planning, and control strategies.

Artificial Intelligence for Transportation, 2025 | 📄 Paper

Fast and Effective Early Termination for Simple Ranking Functions

Jinrui Gou, Antonio Mallia, Yifan Liu, Minghao Shao, Torsten Suel

A novel approach for accelerating web search engines through optimized early termination techniques for simple ranking functions, achieving significant speed improvements while maintaining high retrieval quality on MSMarco V1 and V2 datasets.

ACM SIGIR, 2025 | 📄 Paper

Driving with Regulation: Interpretable Decision-Making for Autonomous Vehicles with Retrieval-Augmented Reasoning via LLM

Tianhui Cai, Yifan Liu, Zewei Zhou, Haoxuan Ma, Seth Z Zhao, Zhiwen Wu, Jiaqi Ma

An interpretable decision-making framework for autonomous vehicles that integrates a Traffic Regulation Retrieval (TRR) Agent using RAG with LLM-powered reasoning to automatically retrieve and interpret traffic rules for legal compliance and safety assessment across different regions.

arXiv preprint, 2024 | 📄 Paper

Beyond Quantile Methods: Improved Top-K Threshold Estimation for Traditional and Learned Sparse Indexes

Jinrui Gou, Yifan Liu, Minghao Shao, Torsten Suel

Enhanced quantile-based methods for top-k threshold estimation that improve upon state-of-the-art approaches in terms of mean under-prediction fraction (MUF) and extend to recently proposed learned sparse index structures for better query processing performance.

IEEE International Conference on Big Data (BigData), 2024 | 📄 Paper

Mobility AI Agents and Networks

Haoxuan Ma, Yifan Liu, Qinhua Jiang, Brian Yueshuai He, Xishun Liao, Jiaqi Ma

An Intelligent Mobility System Digital Twin (MSDT) framework that maps humans and vehicles to AI agents and mobility systems to AI networks, enabling unprecedented prediction and automated simulation capabilities for smart city planning and operations.

IEEE Transactions on Intelligent Vehicles, 2024 | 📄 Paper

Reconstructing Human Mobility Pattern: A Semi-Supervised Approach for Cross-Dataset Transfer Learning

Xishun Liao, Yifan Liu, Chenchen Kuai, Haoxuan Ma, Yueshuai He, Shangqing Cao, Chris Stanford, Jiaqi Ma

A semi-supervised iterative transfer learning algorithm that reconstructs human mobility patterns through semantic activity chains, achieving cross-geographical adaptation with 64% similarity improvement (JSD reduction from 0.09 to 0.03).

arXiv preprint, 2025 | 📄 Paper

Deep Activity Model: A Generative Approach for Human Mobility Pattern Synthesis

Xishun Liao, Qinhua Jiang, Brian Yueshuai He, Yifan Liu, Chenchen Kuai, Jiaqi Ma

A novel generative deep learning framework for synthesizing realistic human mobility patterns, enabling comprehensive modeling of activity sequences and spatial-temporal dynamics for transportation planning and urban analysis applications.

Preprint, 2024 | 📄 Paper

Semantic Trajectory Data Mining with LLM-Informed POI Classification

Yifan Liu, Chenchen Kuai, Haoxuan Ma, Xishun Liao, Brian Yueshuai He, Jiaqi Ma

A novel approach to trajectory data mining that leverages large language models for enhanced point-of-interest (POI) classification, enabling more accurate semantic understanding of mobility patterns and location contexts in transportation systems. 🏆 ITSC 2024 Best Paper Award

IEEE ITSC, 2024 | 📄 Paper

Improved Top-K Threshold Estimation in Information Retrieval

Yifan Liu

In the era of exponential information growth on the internet, the efficiency of search engines has become a critical concern. This thesis addresses the need to enhance the speed and accuracy of search engines by developing improved top-k threshold estimation techniques for information retrieval systems.

New York University Tandon School of Engineering, 2024 | 📄 Paper

Towards Knowledge-Graph Constrained Generation for Text Adventure Games

Graham Todd, Zegang Cheng, Yifan Liu, Julian Togelius

A novel approach to text adventure game generation that leverages knowledge graphs to ensure coherent and logical narrative progression, combining natural language generation with structured world knowledge for enhanced gameplay experiences.

NAACL 2022 The Third Wordplay Workshop, 2022 | 📄 Paper

Does COO Matter in Value Co-creation of Cross-border E-commerce?

Zhen Qin, Yan Ni, Yifan Liu

An empirical study investigating the impact of Country of Origin (COO) effects on value co-creation processes in cross-border e-commerce platforms, exploring consumer behavior and trust dynamics in international online markets.

WHICEB, 2020 | 📄 Paper

Research on the Use of YOLOv5 Object Detection Algorithm in Mask Wearing Recognition

Yifan Liu, BingHang Lu, Jingyu Peng, Zihao Zhang

A comprehensive study on applying YOLOv5 object detection algorithm for automatic mask wearing recognition, addressing the critical need for COVID-19 safety compliance monitoring in public spaces.

World Scientific Research Journal, 2020 | 📄 Paper

🎓 Education

• 2024 - Present, Ph.D., Civil and Environmental Engineering, University of California, Los Angeles (UCLA), Los Angeles, CA
• 2022 - 2024, M.S., Computer Science, New York University (NYU), New York, NY
• 2018 - 2022, B.E., Computer Science and Engineering, University of California, Davis (UC Davis), Davis, CA

🏆 Honors and Awards

• 2025.01 Tier 1 Winner of U.S. DOT Intersection Safety Challenge with a $750k award
• 2024.09 Best Paper Award at IEEE ITSC 2024
• 2021 Dean’s List, UC Davis
• 2020 Outstanding Undergraduate Researcher, UC Davis

📚 Academic Services

Conference Reviewer:

• IEEE Intelligent Transportation Systems Conference (ITSC)
• Transportation Research Board (TRB) Annual Meeting
• IEEE International Conference on Robotics and Systems (IROS)

Journal Reviewer:

• IEEE Open Journal of Intelligent Transportation Systems (OJ-ITS)

Professional Memberships:

• IEEE Member

Workshop Organizer:

• DLIVA Workshop @ IEEE IV 2025
• DriveX Workshop @ ICCV 2025

🎮 Miscellaneous

• I’m an avid yo-yo player and have been practicing for over 8 years - it helps me think creatively about complex problems!
• I enjoy competitive programming and have participated in various coding competitions.
• In my free time, I love hiking in the beautiful mountains around Los Angeles and photography.

Last updated: July 2025