nD-RoPE: A Generalized RoPE for n-Dimensional Position Embedding
Boyang Li, Yulin Wu, Sizhe Xu, Nuoxian Huang, Zhonghang Yuan, Shangyi Guo, Shu Yang, and Takahiro Yabe.
ICML 2026
Boyang Li
Ph.D. Researcher, New York University
Human-Centered Technology, Innovation & Design
I study how humans and intelligent agents perceive, represent, reason, and decide in spatial and urban environments. My research spans spatial cognition, urban AI, representation learning, and collective behavior.
My work connects geospatial machine learning, behavioral decision modeling, causal inference, and computational social science to build interpretable models of how people and artificial agents perceive, choose, and coordinate in space.
Research Interests
-
Spatial Decision-Making & Brain-Inspired Spatial Intelligence
Modeling cognitive maps, route choice, revealed preferences, and spatial decision structures through inverse reinforcement learning and topology-aware AI.
-
Urban Complex Systems & Agentic Collective Intelligence
Designing agent-based and AI-native frameworks for mobility, economic behavior, dependency networks, and emergent coordination in cities.
-
Representation Learning for Spatial AI
Developing general spatial representations for AI systems, including n-dimensional positional encoding and geometry-aware embeddings across coordinates, images, videos, point clouds, and trajectories.
-
Spatial Reasoning & World Models
Building spatial world models for intelligent systems, enabling large language models to understand geographic knowledge, reason over spatial relations, and act within structured environments.
Publications
* Equal contribution · † Corresponding author
LDP-DiffTraj: Diffusion Models under Local Differential Privacy for Personalized Human Mobility Trajectory Generation
Boyang Li, Shangyi Guo, Kota Tsubouchi, Toru Shimizu, Zhonghang Yuan, Joseph Y. J. Chow, and Takahiro Yabe.
ACM SIGSPATIAL 2026 · Accepted · Proceedings forthcoming
Best Paper Candidate
Behaviorally Grounding Large Language Models with Revealed Choices for Human Mobility Prediction
Boyang Li*, Sizhe Xu*, DongHak Lee, Shangyi Guo, Zhaonan Wang, and Takahiro Yabe.
ACM SIGSPATIAL 2026 · Accepted · Proceedings forthcoming
Enhancing Pedestrian Route Choice Models Through Maximum-Entropy Deep Inverse Reinforcement Learning With Individual Covariates (MEDIRL-IC)
Boyang Li and Wenjia Zhang.
IEEE Transactions on Intelligent Transportation Systems · 2024
GridPE: A Grid Cell-Inspired Unified Position Embedding for Arbitrary-Dimensional Spaces
Boyang Li, Yulin Wu, Nuoxian Huang, and Wenjia Zhang.
arXiv · 2024 · precursor to nD-RoPE
AskNearby: An LLM-Based Application for Neighborhood Information Retrieval and Personalized Cognitive-Map Recommendations
Luyao Niu, Zhicheng Deng, Boyang Li†, Nuoxian Huang, Ruiqi Liu, and Wenjia Zhang.
ACM SIGSPATIAL GeoAI Workshop · 2025
Varying Built Environment Contexts and Trip Chain Decisions: A Multinomial-Choice Gradient Boosting Decision Trees Analysis
Wenjia Zhang, Daming Lu, Hongjin Liu, and Boyang Li.
Travel Behaviour and Society · 2024
PlanGPT: Enhancing Urban Planning with Tailored Language Model and Efficient Retrieval
He Zhu, Wenjia Zhang, Nuoxian Huang, Boyang Li, et al.
arXiv · 2024
Education
- New York University, Ph.D. in Human-Centered Technology, Innovation & Design, 2025-present.
- Peking University, M.Sc. in Geography, Smart Cities and Big Data, 2022-2025.
- Wuhan University, B.Eng. in Remote Sensing Science and Technology, 2018-2022.
Grants & Fellowships
- NYU Tandon School of Engineering Ph.D. Fellowship, fully funded (2025-2026).
- ACM SIGSPATIAL 2026 Student Travel Grant, $500 (2026).
- Yi-Innovator College Student Social Innovation Grant (2024).
- Baidu Wenxin Cup Research Scholarship (2024).
- Peking University Long Yuan Chang Ze Scholarship (2024).
- Peking University Shenzhen Graduate School Tie Han Scholarship (2023).
- Wuhan University First-Class Scholarship (2019 · 2021).
- Wuhan University Hong Tu Chuang Zhan Scholarship (2020).
Contact
Open to research conversations on spatial intelligence and urban AI.