# ReasonLight: A Multimodal Foundation Model-Enhanced Reinforcement Learning Framework for Zero-Shot Traffic Signal Control

arXiv: https://arxiv.org/abs/2605.29425 · HTML: https://mlsanity.com/paper/2605.29425/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for autonomous driving and traffic, one of 15 directions in Deep reinforcement learning methods.
- Builds on 21 of its 40 references that are in the map.
- First posted on arXiv 28 May 2026.

## Placed in

- Deep reinforcement learning methods / [RL for autonomous driving and traffic](https://mlsanity.com/deeprl/rl-for-autonomous-driving-and-traffic/)

## Builds on (most cited first, 21 of 21)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [A Survey on Multimodal Large Language Models](https://mlsanity.com/paper/2306.13549/) (2023)
- [Multimodal Foundation Models: From Specialists to General-Purpose Assistants](https://mlsanity.com/paper/2309.10020/) (2023)
- [Improve Vision Language Model Chain-of-thought Reasoning](https://mlsanity.com/paper/2410.16198/) (2024)
- [VLM-AD: End-to-End Autonomous Driving through Vision-Language Model Supervision](https://mlsanity.com/paper/2412.14446/) (2024)
- [A Survey on Traffic Signal Control Methods](https://mlsanity.com/paper/1904.08117/) (2019)
- [LLMLight: Large Language Models as Traffic Signal Control Agents](https://mlsanity.com/paper/2312.16044/) (2023)
- [AttendLight: Universal Attention-Based Reinforcement Learning Model for Traffic Signal Control](https://mlsanity.com/paper/2010.05772/) (2020)
- [LLM-Assisted Light: Leveraging Large Language Model Capabilities for Human-Mimetic Traffic Signal Control in Complex Urban Environments](https://mlsanity.com/paper/2403.08337/) (2024)
- [iLLM-TSC: Integration reinforcement learning and large language model for traffic signal control policy improvement](https://mlsanity.com/paper/2407.06025/) (2024)
- [CoLLMLight: Cooperative Large Language Model Agents for Network-Wide Traffic Signal Control](https://mlsanity.com/paper/2503.11739/) (2025)
- [TrafficLLM: Enhancing Large Language Models for Network Traffic Analysis with Generic Traffic Representation](https://mlsanity.com/paper/2504.04222/) (2025)
- [UniTSA: A Universal Reinforcement Learning Framework for V2X Traffic Signal Control](https://mlsanity.com/paper/2312.05090/) (2023)
- [Traffic-R1: Reinforced LLMs Bring Human-Like Reasoning to Traffic Signal Control Systems](https://mlsanity.com/paper/2508.02344/) (2025)
- [A Decentralized Reinforcement Learning Framework for Efficient Passage of Emergency Vehicles](https://mlsanity.com/paper/2111.00278/) (2021)
- [Traffic Signal Cycle Control with Centralized Critic and Decentralized Actors under Varying Intervention Frequencies](https://mlsanity.com/paper/2406.08248/) (2024)
- [Traffic Modeling with SUMO: a Tutorial](https://mlsanity.com/paper/2304.05982/) (2023)
- [VLMLight: Safety-Critical Traffic Signal Control via Vision-Language Meta-Control and Dual-Branch Reasoning Architecture](https://mlsanity.com/paper/2505.19486/) (2025)
- [TranSimHub:A Unified Air-Ground Simulation Platform for Multi-Modal Perception and Decision-Making](https://mlsanity.com/paper/2510.15365/) (2025)

## Abstract (arXiv)

Reinforcement learning (RL) has shown promise in traffic signal control (TSC). However, its reliance on predefined states limits responsiveness to observable open-world events that are absent from training data. IoT-enabled intersections provide heterogeneous observations from roadside sensors and cameras, creating opportunities to improve RL adaptability to such events. To this end, we propose ReasonLight, a multimodal foundation model-enhanced RL framework for zero-shot TSC. ReasonLight integrates three sources of information: structured traffic measurements, multi-view camera observations, and candidate phase decisions from a pre-trained RL controller. Given an RL-proposed phase, ReasonLight extracts visual semantics from multi-view images and aligns them with compact sensor-derived scene descriptions. This alignment enables a semantic-guided refinement module to either preserve or adjust the proposed action according to traffic rules and event semantics. To ensure operational reliability, refined actions are constrained by the set of available phases. Any invalid decision is rejected, and the system falls back to the original RL action. We evaluate ReasonLight on two types of rare events not seen during RL training: emergency vehicle priority and temporary traffic regulation. Experimental results show that ReasonLight achieves zero-shot adaptation without retraining. It reduces emergency vehicle waiting time by up to 88.7% compared with the RL-only backbone while preserving comparable routine traffic performance.