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

• URL: http://arxiv.org/abs/2605.29425v1
• Date: Thu, 28 May 2026 06:19:09 GMT
• Title: ReasonLight: A Multimodal Foundation Model-Enhanced Reinforcement Learning Framework for Zero-Shot Traffic Signal Control
• Abstract summary: ReasonLight is a multimodal foundation model-enhanced RL framework for zero-shot TSC.<n>It integrates structured traffic measurements, multi-view camera observations, and candidate phase decisions from a pre-trained RL controller.<n>It reduces emergency vehicle waiting time by up to 88.7% compared with the RL-only backbone.
• Score: 7.396142955449211
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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.

Related papers

• COVLM-RL: Critical Object-Oriented Reasoning for Autonomous Driving Using VLM-Guided Reinforcement Learning [55.83415345423854]
  We propose COVLM-RL, a novel end-to-end driving framework that integrates Critical Object-oriented (CO) reasoning with reinforcement learning (RL)<n>Experiments conducted in the CARLA simulator demonstrate that COVLM-RL significantly improves the success rate by 30% in trained driving environments.
  arXiv  Detail & Related papers  (2025-12-10T06:18:16Z)
• The Path Not Taken: RLVR Provably Learns Off the Principals [85.41043469428365]
  We show that sparsity is a surface artifact of a model-conditioned optimization bias.<n>We mechanistically explain these dynamics with a Three-Gate Theory.<n>We provide a parameter-level characterization of RLVR's learning dynamics.
  arXiv  Detail & Related papers  (2025-11-11T18:49:45Z)
• Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning [93.00629872970364]
  Reinforcement learning (RL) has become the dominant paradigm for improving the performance of language models on complex reasoning tasks.<n>We introduce SPARKLE, a fine-grained analytic framework to dissect the effects of RL across three key dimensions.<n>We study whether difficult problems -- those yielding no RL signals and mixed-quality reasoning traces -- can still be effectively used for training.
  arXiv  Detail & Related papers  (2025-06-05T07:53:59Z)
• VLMLight: Safety-Critical Traffic Signal Control via Vision-Language Meta-Control and Dual-Branch Reasoning Architecture [10.403439446065033]
  VLMLight is a novel framework that integrates vision-language meta-control with dual-branch reasoning.<n>A large language model (LLM) serves as a safety-prioritized meta-controller, selecting between a fast RL policy for routine traffic and a structured reasoning branch for critical cases.<n> Experiments show that VLMLight reduces waiting times for emergency vehicles by up to 65% over RL-only systems.
  arXiv  Detail & Related papers  (2025-05-26T04:12:57Z)
• ViaRL: Adaptive Temporal Grounding via Visual Iterated Amplification Reinforcement Learning [68.76048244253582]
  We introduce ViaRL, the first framework to leverage rule-based reinforcement learning (RL) for optimizing frame selection in video understanding.<n>ViaRL utilizes the answer accuracy of a downstream model as a reward signal to train a frame selector through trial-and-error.<n>ViaRL consistently delivers superior temporal grounding performance and robust generalization across diverse video understanding tasks.
  arXiv  Detail & Related papers  (2025-05-21T12:29:40Z)
• FitLight: Federated Imitation Learning for Plug-and-Play Autonomous Traffic Signal Control [33.547772623142414]
  Reinforcement Learning (RL)-based Traffic Signal Control (TSC) methods raise some serious issues such as high learning cost and poor generalizability.<n>We propose a novel Federated Imitation Learning (FIL)-based framework for multi-intersection TSC, named FitLight.<n> FitLight allows real-time imitation learning and seamless transition to reinforcement learning.
  arXiv  Detail & Related papers  (2025-02-17T15:48:46Z)
• iLLM-TSC: Integration reinforcement learning and large language model for traffic signal control policy improvement [5.078593258867346]
  We introduce a novel integration framework that combines a large language model (LLM) with reinforcement learning (RL) Our approach reduces the average waiting time by $17.5%$ in degraded communication conditions as compared to traditional RL methods.
  arXiv  Detail & Related papers  (2024-07-08T15:22:49Z)
• A Holistic Framework Towards Vision-based Traffic Signal Control with Microscopic Simulation [53.39174966020085]
  Traffic signal control (TSC) is crucial for reducing traffic congestion that leads to smoother traffic flow, reduced idling time, and mitigated CO2 emissions. In this study, we explore the computer vision approach for TSC that modulates on-road traffic flows through visual observation. We introduce a holistic traffic simulation framework called TrafficDojo towards vision-based TSC and its benchmarking.
  arXiv  Detail & Related papers  (2024-03-11T16:42:29Z)
• DenseLight: Efficient Control for Large-scale Traffic Signals with Dense Feedback [109.84667902348498]
  Traffic Signal Control (TSC) aims to reduce the average travel time of vehicles in a road network. Most prior TSC methods leverage deep reinforcement learning to search for a control policy. We propose DenseLight, a novel RL-based TSC method that employs an unbiased reward function to provide dense feedback on policy effectiveness.
  arXiv  Detail & Related papers  (2023-06-13T05:58:57Z)
• Reinforcement Learning Approaches for Traffic Signal Control under Missing Data [5.896742981602458]
  In real-world urban scenarios, missing observation of traffic states may frequently occur due to the lack of sensors. We propose two solutions: the first one imputes the traffic states to enable adaptive control, and the second one imputes both states and rewards to enable adaptive control and the training of RL agents.
  arXiv  Detail & Related papers  (2023-04-21T03:26:33Z)
• ModelLight: Model-Based Meta-Reinforcement Learning for Traffic Signal Control [5.219291917441908]
  This paper proposes a novel model-based meta-reinforcement learning framework (ModelLight) for traffic signal control. Within ModelLight, an ensemble of models for road intersections and the optimization-based meta-learning method are used to improve the data efficiency of an RL-based traffic light control method. Experiments on real-world datasets demonstrate that ModelLight can outperform state-of-the-art traffic light control algorithms.
  arXiv  Detail & Related papers  (2021-11-15T20:25:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.