Bridging Policy and Real-World Dynamics: LLM-Augmented Rebalancing for Shared Micromobility Systems

• URL: http://arxiv.org/abs/2603.00176v1
• Date: Thu, 26 Feb 2026 19:13:17 GMT
• Title: Bridging Policy and Real-World Dynamics: LLM-Augmented Rebalancing for Shared Micromobility Systems
• Abstract summary: We introduce AMPLIFY, an LLM-augmented policy adaptation framework for shared micromobility rebalancing.<n>The framework combines a baseline rebalancing module with an LLM-based adaptation module that adjusts strategies in real time under emergent scenarios.<n> Evaluations on real-world e-scooter data from Chicago show that our approach improves demand satisfaction and system revenue compared to baseline policies.
• Score: 8.57514610822263
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Shared micromobility services such as e-scooters and bikes have become an integral part of urban transportation, yet their efficiency critically depends on effective vehicle rebalancing. Existing methods either optimize for average demand patterns or employ robust optimization and reinforcement learning to handle predefined uncertainties. However, these approaches overlook emergent events (e.g., demand surges, vehicle outages, regulatory interventions) or sacrifice performance in normal conditions. We introduce AMPLIFY, an LLM-augmented policy adaptation framework for shared micromobility rebalancing. The framework combines a baseline rebalancing module with an LLM-based adaptation module that adjusts strategies in real time under emergent scenarios. The adaptation module ingests system context, demand predictions, and baseline strategies, and refines adjustments through self-reflection. Evaluations on real-world e-scooter data from Chicago show that our approach improves demand satisfaction and system revenue compared to baseline policies, highlighting the potential of LLM-driven adaptation as a flexible solution for managing uncertainty in micromobility systems.

Related papers

• CoRAL: Contact-Rich Adaptive LLM-based Control for Robotic Manipulation [4.600772894287281]
  Large Language Models (LLMs) and Vision-Language Models (VLMs) demonstrate remarkable capabilities in high-level reasoning and semantic understanding.<n>We propose CoRAL, a modular framework that enables zero-shot planning by decoupling high-level reasoning from low-level control.
  arXiv  Detail & Related papers  (2026-05-04T13:49:19Z)
• MAESTRO: Meta-learning Adaptive Estimation of Scalarization Trade-offs for Reward Optimization [56.074760766965085]
  Group-Relative Policy Optimization has emerged as an efficient paradigm for aligning Large Language Models (LLMs)<n>We propose MAESTRO, which treats reward scalarization as a dynamic latent policy, leveraging the model's terminal hidden states as a semantic bottleneck.<n>We formulate this as a contextual bandit problem within a bi-level optimization framework, where a lightweight Conductor network co-evolves with the policy by utilizing group-relative advantages as a meta-reward signal.
  arXiv  Detail & Related papers  (2026-01-12T05:02:48Z)
• BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping [69.74252624161652]
  We propose BAlanced Policy Optimization with Adaptive Clipping (BAPO)<n>BAPO dynamically adjusts clipping bounds to adaptively re-balance positive and negative contributions, preserve entropy, and stabilize RL optimization.<n>On AIME 2024 and AIME 2025 benchmarks, our 7B BAPO model surpasses open-source counterparts such as SkyWork-OR1-7B.
  arXiv  Detail & Related papers  (2025-10-21T12:55:04Z)
• Hierarchical Optimization via LLM-Guided Objective Evolution for Mobility-on-Demand Systems [9.979671028876464]
  We propose a novel framework that integrates large language model (LLM) with mathematical optimization in a dynamic hierarchical system.<n>Within this framework, LLM serves as a meta-optimizer, producing semantics that guide a low-level responsible for constraint enforcement and real-time decision execution.<n>Experiments based on both the New York and Chicago taxi datasets demonstrate the effectiveness of our approach.
  arXiv  Detail & Related papers  (2025-10-12T14:56:19Z)
• Adaptive Composition of Machine Learning as a Service (MLaaS) for IoT Environments [0.0]
  The dynamic nature of the Internet of Things (IoT) environments challenges the effectiveness of Machine Learning as a Service (ML) compositions.<n>This paper proposes an adaptive ML composition framework to ensure a seamless, efficient, and scalable ML composition.
  arXiv  Detail & Related papers  (2025-05-22T11:31:00Z)
• Modality-Balancing Preference Optimization of Large Multimodal Models by Adversarial Negative Mining [75.14823970163685]
  We propose a novel preference learning framework, Modality-Balancing Preference Optimization (MBPO), to address the modality imbalance in LMMs.<n>MBPO constructs a more effective offline preference dataset by generating hard negatives, i.e., rejected responses misled by LLM biases.<n>It can enhance LMM performance on challenging vision-language tasks and effectively reduce hallucinations.
  arXiv  Detail & Related papers  (2025-05-20T03:59:05Z)
• Dynamic Rewarding with Prompt Optimization Enables Tuning-free Self-Alignment of Language Models [54.381650481255235]
  We introduce a new tuning-free approach for self-alignment, Dynamic Rewarding with Prompt Optimization (O) Our approach leverages a search-based optimization framework that allows LLMs to iteratively self-improve and craft the optimal alignment instructions. Empirical evaluations on eight recent LLMs, both open and closed-sourced, demonstrate that DRPO significantly enhances alignment performance.
  arXiv  Detail & Related papers  (2024-11-13T16:15:38Z)
• On-the-fly Modulation for Balanced Multimodal Learning [53.616094855778954]
  Multimodal learning is expected to boost model performance by integrating information from different modalities. The widely-used joint training strategy leads to imbalanced and under-optimized uni-modal representations. We propose On-the-fly Prediction Modulation (OPM) and On-the-fly Gradient Modulation (OGM) strategies to modulate the optimization of each modality.
  arXiv  Detail & Related papers  (2024-10-15T13:15:50Z)
• Attribute Controlled Fine-tuning for Large Language Models: A Case Study on Detoxification [76.14641982122696]
  We propose a constraint learning schema for fine-tuning Large Language Models (LLMs) with attribute control. We show that our approach leads to an LLM that produces fewer inappropriate responses while achieving competitive performance on benchmarks and a toxicity detection task.
  arXiv  Detail & Related papers  (2024-10-07T23:38:58Z)
• Reinforcement Learning with Model Predictive Control for Highway Ramp Metering [14.389086937116582]
  This work explores the synergy between model-based and learning-based strategies to enhance traffic flow management.<n>The control problem is formulated as an RL task by crafting a suitable stage cost function.<n>An MPC-based RL approach, which leverages the MPC optimal problem as a function approximation for the RL algorithm, is proposed to learn to efficiently control an on-ramp.
  arXiv  Detail & Related papers  (2023-11-15T09:50:54Z)
• Towards More Efficient Shared Autonomous Mobility: A Learning-Based Fleet Repositioning Approach [0.0]
  This paper formulates SAMS fleet as a Markov Decision Process and presents a reinforcement learning-based repositioning (RLR) approach called integrated system-agent repositioning (ISR) The ISR learns to respond to evolving demand patterns without explicit demand forecasting and to cooperate with optimization-based passenger-to-vehicle assignment. Results demonstrate the RLR approaches' substantial reductions in passenger wait times, over 50%, relative to the JO approach.
  arXiv  Detail & Related papers  (2022-10-16T23:30:46Z)
• A Modular and Transferable Reinforcement Learning Framework for the Fleet Rebalancing Problem [2.299872239734834]
  We propose a modular framework for fleet rebalancing based on model-free reinforcement learning (RL) We formulate RL state and action spaces as distributions over a grid of the operating area, making the framework scalable. Numerical experiments, using real-world trip and network data, demonstrate that this approach has several distinct advantages over baseline methods.
  arXiv  Detail & Related papers  (2021-05-27T16:32:28Z)

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.