activity
20242026
collaborators

19 papers

cs.LG2026

Forecasting the Emergence and Evolution of Crash Hotspots: A Unified Deep Learning Framework for Proactive Traffic Safety

Jingwen Zhu, Keshu Wu, Pei Li +3

Road crashes remain among the gravest threats to public safety, and preventing them is a defining task of transportation systems worldwide. Much of that harm concentrates at hotspo…

cs.CV2026

SUMO: Segment and Track Any Motion with Nonlinear State Space Models

Kexin Tian, Sixu Li, Keshu Wu +2

Visual Object Tracking (VOT) and Moving Object Segmentation (MOS) are two fundamental tasks in computer vision that involve both spatial and temporal object dynamics. Existing meth…

cs.LG2026

Learning Higher-Order Structure from Incomplete Spatiotemporal Data: Multi-Scale Hypergraph Laplacians with Neural Refinement

Keshu Wu, Sixu Li, Zihao Li +3

Sensor networks increasingly govern modern infrastructure, yet the data they lose are rarely missing in the uniform-random patterns assumed by standard imputation benchmarks. Loop…

cs.CL2026

Knowledge Is Not Static: Order-Aware Hypergraph RAG for Language Models

Keshu Wu, Chenchen Kuai, Zihao Li +6

Retrieval-augmented generation (RAG) enhances large language models by grounding outputs in retrieved knowledge. However, existing RAG methods including graph- and hypergraph-based…

cs.AI2026

A Statistical Framework for Auditing Behavioral Dependence and Induced Bias in LLM Judges

Chenchen Kuai, Jiwan Jiang, Zihao Zhu +8

The rapid growth of the large language model (LLM) ecosystem raises a critical question: are seemingly diverse models truly independent? Shared pretraining data, distillation, and…

cs.CE2025

Human Mobility Reimagined: Digital Twin Intelligence for Adaptive Campus Course Timetabling

Keshu Wu, Xinyue Ye, Suphanut Jamonnak +1

Daily operations in large campuses depend on how efficiently people \emph{move} through space and time. In this sense, course timetables are more than administrative schedules: the…