mlsanityData to 14 Oct 2026

CROSS: A Mixture-of-Experts Reinforcement Learning Framework for Generalizable Large-Scale Traffic Signal Control

arXiv 2603.24930 ↗First posted 26 Mar 2026cs.ROcited by 0

On the map: Deep reinforcement learning methods → RL for autonomous driving and traffic

Abstract (arXiv)

Recent advances in robotics, automation, and artificial intelligence have enabled urban traffic systems to operate with increasing autonomy towards future smart cities, powered in part by the development of adaptive traffic signal control (ATSC), which dynamically optimizes signal phases to mitigate congestion and optimize traffic. However, achieving effective and generalizable large-scale ATSC remains a significant challenge due to the diverse intersection topologies and highly dynamic, complex traffic demand patterns across the network. Existing RL-based methods typically use a single shared policy for all scenarios, whose limited representational capacity makes it difficult to capture diverse traffic dynamics and generalize to unseen environments. To address these challenges, we propose CROSS, a novel Mixture-of-Experts (MoE)-based decentralized RL framework for generalizable ATSC. We first introduce a Predictive Contrastive Clustering (PCC) module that forecasts short-term state transitions to identify latent traffic patterns, followed by clustering and contrastive learning to enhance pattern-level representation. We further design a Scenario-Adaptive MoE module that augments a shared policy with multiple experts, thus enabling adaptive specialization and more flexible scenario-specific strategies. We conduct extensive experiments in the SUMO simulator on both synthetic and real-world traffic datasets. Compared with state-of-the-art baselines, CROSS achieves superior performance and generalization through improved representation of diverse traffic scenarios.

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 16 of its 31 references that are in the map. First posted on arXiv 26 Mar 2026.

Builds on

16 of 31 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Unicorn: A Universal and Collaborative Reinforcement Learning Approach Towards Generalizable Network-Wide Traffic Signal Control14 Mar

Other references · 15

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2017Proximal Policy Optimization Algorithmscited by 15,215
  3. 2018Representation Learning with Contrastive Predictive Codingcited by 7,777
  4. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,367
  5. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 2,026
  6. 2019Multi-Agent Deep Reinforcement Learning for Large-scale Traffic Signal Controlcited by 164
  7. 2019CoLight: Learning Network-level Cooperation for Traffic Signal Controlcited by 138
  8. 2020Deep Reinforcement Learning for Intelligent Transportation Systems: A Surveycited by 98
  9. 2019Learning Phase Competition for Traffic Signal Controlcited by 87
  10. 2020AttendLight: Universal Attention-Based Reinforcement Learning Model for Traffic Signal Controlcited by 62
  11. 2021Expression might be enough: representing pressure and demand for reinforcement learning based traffic signal controlcited by 29
  12. 2023UniTSA: A Universal Reinforcement Learning Framework for V2X Traffic Signal Controlcited by 13
3 more
  1. 2021MetaVIM: Meta Variationally Intrinsic Motivated Reinforcement Learning for Decentralized Traffic Signal Controlcited by 8
  2. 2023SocialLight: Distributed Cooperation Learning towards Network-Wide Traffic Signal Controlcited by 6
  3. 2026CoordLight: Learning Decentralized Coordination for Network-Wide Traffic Signal Controlcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Large-Scale Mixed-Traffic and Intersection Control using Multi-agent Reinforcement Learning
  2. 2024Reinforcement Learning for Adaptive Traffic Signal Control: Turn-Based and Time-Based Approaches to Reduce Congestion
  3. 2025HALO: Hierarchical Reinforcement Learning for Large-Scale Adaptive Traffic Signal Control
  4. 2024Neighbor-Aware Reinforcement Learning for Mixed Traffic Optimization in Large-scale Networks
  5. 2025Federated Hierarchical Reinforcement Learning for Adaptive Traffic Signal Control
  6. 2025Optimizing Efficiency of Mixed Traffic through Reinforcement Learning: A Topology-Independent Approach and Benchmark
  7. 2025Robust Single-Agent Reinforcement Learning for Regional Traffic Signal Control Under Demand Fluctuations
  8. 2025Single-agent Reinforcement Learning Model for Regional Adaptive Traffic Signal Control