TrafficClaw: Generalizable Urban Traffic Control via Unified Physical Environment Modeling

ArXi:2604.17456v1 Announce Type: new Urban traffic control is a system-level coordination problem spanning heterogeneous subsystems, including traffic signals, freeways, public transit, and taxi services. Existing optimization-based, reinforcement learning (RL), and emerging LLM-based approaches are largely designed for isolated tasks, limiting both cross-task generalization and the ability to capture coupled physical dynamics across subsystems.