Guangyin Jin

Multi-Agent Collaborative Reasoning with Tool-Augmented Evidence for Urban Region Profiling

This work proposes UrbanAgent, an agentic framework that reframes urban region profiling as a reasoning-driven inference problem, and extends indicator prediction as a closed-loop process of active evidence acquisition and iterative reasoning, enabling agents to verify uncertain inferences through tool-augmented retrie...

STKAN: Kolmogorov-Arnold Networks for Spatio-Temporal Forecasting

STKAN is proposed, a spatio-temporal forecasting architecture that introduces Taylor-polynomial Kolmogorov--Arnold Network modules into spatial and temporal token mixing and achieves competitive performance and performs better than the evaluated MLP-based variant in the tested settings.

Cost-effective and scalable traffic forecasting: A graph-free multi-view MLP architecture.

The Spatio-Temporal Multi-view MLP Network (STMMN), a highly scalable, graph-free computational framework that introduces a temporal multi-view aggregation module that seamlessly integrates distinct temporal branches, to dynamically fuse multi-scale temporal dependencies without relying on exhaustive attention maps.

Identity-embedded graph neural networks for node-level cost performance index prediction in electrical construction tasks