Language-Driven Policy Distillation for Cooperative Driving in Multi-Agent Reinforcement Learning

The cooperative driving technology of Connected and Autonomous Vehicles (CAVs) is crucial for improving the efficiency and safety of transportation systems. Learning-based methods, such as Multi-Agent Reinforcement Learning (MARL), have demonstrated strong capabilities in cooperative decision-making tasks. However, existing MARL approaches still face challenges in terms of learning efficiency and performance. In recent years, Large Language Models (LLMs) have rapidly advanced and shown remarkable abilities in various sequential decision-making tasks. To enhance the learning capabilities of cooperative agents while ensuring decision-making efficiency and cost-effectiveness, we propose LDPD, a language-driven policy distillation method for guiding MARL exploration. In this framework, a teacher agent based on LLM trains smaller student agents to achieve cooperative decision-making through its own decision-making demonstrations. The teacher agent enhances the observation information of CAVs and utilizes LLMs to perform complex cooperative decision-making reasoning, which also leverages carefully designed decision-making tools to achieve expert-level decisions, providing high-quality teaching experiences. The student agent then refines the teacher's prior knowledge into its own model through gradient policy updates. The experiments demonstrate that the students can rapidly improve their capabilities with minimal guidance from the teacher and eventually surpass the teacher's performance. Extensive experiments show that our approach demonstrates better performance and learning efficiency compared to baseline methods.

Cooperative Decision-Making for CAVs at Unsignalized Intersections: A MARL Approach with Attention and Hierarchical Game Priors

The development of autonomous vehicles has shown great potential to enhance the efficiency and safety of transportation systems. However, the decision-making issue in complex human-machine mixed traffic scenarios, such as unsignalized intersections, remains a challenge for autonomous vehicles. While reinforcement learning (RL) has been used to solve complex decision-making problems, existing RL methods still have limitations in dealing with cooperative decision-making of multiple connected autonomous vehicles (CAVs), ensuring safety during exploration, and simulating realistic human driver behaviors. In this paper, a novel and efficient algorithm, Multi-Agent Game-prior Attention Deep Deterministic Policy Gradient (MA-GA-DDPG), is proposed to address these limitations. Our proposed algorithm formulates the decision-making problem of CAVs at unsignalized intersections as a decentralized multi-agent reinforcement learning problem and incorporates an attention mechanism to capture interaction dependencies between ego CAV and other agents. The attention weights between the ego vehicle and other agents are then used to screen interaction objects and obtain prior hierarchical game relations, based on which a safety inspector module is designed to improve the traffic safety. Furthermore, both simulation and hardware-in-the-loop experiments were conducted, demonstrating that our method outperforms other baseline approaches in terms of driving safety, efficiency, and comfort.

A Survey on Diffusion Models for Recommender Systems

While traditional recommendation techniques have made significant strides in the past decades, they still suffer from limited generalization performance caused by factors like inadequate collaborative signals, weak latent representations, and noisy data. In response, diffusion models (DMs) have emerged as promising solutions for recommender systems due to their robust generative capabilities, solid theoretical foundations, and improved training stability. To this end, in this paper, we present the first comprehensive survey on diffusion models for recommendation, and draw a bird's-eye view from the perspective of the whole pipeline in real-world recommender systems. We systematically categorize existing research works into three primary domains: (1) diffusion for data engineering & encoding, focusing on data augmentation and representation enhancement; (2) diffusion as recommender models, employing diffusion models to directly estimate user preferences and rank items; and (3) diffusion for content presentation, utilizing diffusion models to generate personalized content such as fashion and advertisement creatives. Our taxonomy highlights the unique strengths of diffusion models in capturing complex data distributions and generating high-quality, diverse samples that closely align with user preferences. We also summarize the core characteristics of the adapting diffusion models for recommendation, and further identify key areas for future exploration, which helps establish a roadmap for researchers and practitioners seeking to advance recommender systems through the innovative application of diffusion models. To further facilitate the research community of recommender systems based on diffusion models, we actively maintain a GitHub repository for papers and other related resources in this rising direction https://github.com/CHIANGEL/Awesome-Diffusion-for-RecSys.

Towards Safe and Robust Autonomous Vehicle Platooning: A Self-Organizing Cooperative Control Framework

In the emerging hybrid traffic flow environment, which includes both human-driven vehicles (HDVs) and autonomous vehicles (AVs), ensuring safe and robust decision-making and control is crucial for the effective operation of autonomous vehicle platooning. Current systems for cooperative adaptive cruise control and lane changing are inadequate in responding to real-world emergency situations, limiting the potential of autonomous vehicle platooning technology. To address the aforementioned challenges, we propose a Twin-World Safety-Enhanced Data-Model-Knowledge Hybrid-Driven autonomous vehicle platooning Cooperative Control Framework. Within this framework, a deep reinforcement learning formation decision model integrating traffic priors is designed, and a twin-world deduction model based on safety priority judgment is proposed. Subsequently, an optimal control-based multi-scenario decision-control right adaptive switching mechanism is designed to achieve adaptive switching between data-driven and model-driven methods. Through simulation experiments and hardware-in-loop tests, our algorithm has demonstrated excellent performance in terms of safety, robustness, and flexibility. A detailed account of the validation results for the model can be found in \url{https://perfectxu88.github.io/towardssafeandrobust.github.io/}.

MAPPO-PIS: A Multi-Agent Proximal Policy Optimization Method with Prior Intent Sharing for CAVs' Cooperative Decision-Making

Vehicle-to-Vehicle (V2V) technologies have great potential for enhancing traffic flow efficiency and safety. However, cooperative decision-making in multi-agent systems, particularly in complex human-machine mixed merging areas, remains challenging for connected and autonomous vehicles (CAVs). Intent sharing, a key aspect of human coordination, may offer an effective solution to these decision-making problems, but its application in CAVs is under-explored. This paper presents an intent-sharing-based cooperative method, the Multi-Agent Proximal Policy Optimization with Prior Intent Sharing (MAPPO-PIS), which models the CAV cooperative decision-making problem as a Multi-Agent Reinforcement Learning (MARL) problem. It involves training and updating the agents' policies through the integration of two key modules: the Intention Generator Module (IGM) and the Safety Enhanced Module (SEM). The IGM is specifically crafted to generate and disseminate CAVs' intended trajectories spanning multiple future time-steps. On the other hand, the SEM serves a crucial role in assessing the safety of the decisions made and rectifying them if necessary. Merging area with human-machine mixed traffic flow is selected to validate our method. Results show that MAPPO-PIS significantly improves decision-making performance in multi-agent systems, surpassing state-of-the-art baselines in safety, efficiency, and overall traffic system performance. The code and video demo can be found at: \url{https://github.com/CCCC1dhcgd/A-MAPPO-PIS}.

Training-free Composite Scene Generation for Layout-to-Image Synthesis

Recent breakthroughs in text-to-image diffusion models have significantly advanced the generation of high-fidelity, photo-realistic images from textual descriptions. Yet, these models often struggle with interpreting spatial arrangements from text, hindering their ability to produce images with precise spatial configurations. To bridge this gap, layout-to-image generation has emerged as a promising direction. However, training-based approaches are limited by the need for extensively annotated datasets, leading to high data acquisition costs and a constrained conceptual scope. Conversely, training-free methods face challenges in accurately locating and generating semantically similar objects within complex compositions. This paper introduces a novel training-free approach designed to overcome adversarial semantic intersections during the diffusion conditioning phase. By refining intra-token loss with selective sampling and enhancing the diffusion process with attention redistribution, we propose two innovative constraints: 1) an inter-token constraint that resolves token conflicts to ensure accurate concept synthesis; and 2) a self-attention constraint that improves pixel-to-pixel relationships. Our evaluations confirm the effectiveness of leveraging layout information for guiding the diffusion process, generating content-rich images with enhanced fidelity and complexity. Code is available at https://github.com/Papple-F/csg.git.

A Decision-Making GPT Model Augmented with Entropy Regularization for Autonomous Vehicles

In the domain of autonomous vehicles (AVs), decision-making is a critical factor that significantly influences the efficacy of autonomous navigation. As the field progresses, the enhancement of decision-making capabilities in complex environments has become a central area of research within data-driven methodologies. Despite notable advances, existing learning-based decision-making strategies in autonomous vehicles continue to reveal opportunities for further refinement, particularly in the articulation of policies and the assurance of safety. In this study, the decision-making challenges associated with autonomous vehicles are conceptualized through the framework of the Constrained Markov Decision Process (CMDP) and approached as a sequence modeling problem. Utilizing the Generative Pre-trained Transformer (GPT), we introduce a novel decision-making model tailored for AVs, which incorporates entropy regularization techniques to bolster exploration and enhance safety performance. Comprehensive experiments conducted across various scenarios affirm that our approach surpasses several established baseline methods, particularly in terms of safety and overall efficacy.

RGB-T Object Detection via Group Shuffled Multi-receptive Attention and Multi-modal Supervision

Multispectral object detection, utilizing both visible (RGB) and thermal infrared (T) modals, has garnered significant attention for its robust performance across diverse weather and lighting conditions. However, effectively exploiting the complementarity between RGB-T modals while maintaining efficiency remains a critical challenge. In this paper, a very simple Group Shuffled Multi-receptive Attention (GSMA) module is proposed to extract and combine multi-scale RGB and thermal features. Then, the extracted multi-modal features are directly integrated with a multi-level path aggregation neck, which significantly improves the fusion effect and efficiency. Meanwhile, multi-modal object detection often adopts union annotations for both modals. This kind of supervision is not sufficient and unfair, since objects observed in one modal may not be seen in the other modal. To solve this issue, Multi-modal Supervision (MS) is proposed to sufficiently supervise RGB-T object detection. Comprehensive experiments on two challenging benchmarks, KAIST and DroneVehicle, demonstrate the proposed model achieves the state-of-the-art accuracy while maintaining competitive efficiency.

Evaluation of Drivers' Interaction Ability at Social Scenarios: A Process-Based Framework

Assessing drivers' interaction capabilities is crucial for understanding human driving behavior and enhancing the interactive abilities of autonomous vehicles. In scenarios involving strong interaction, existing metrics focused on interaction outcomes struggle to capture the evolutionary process of drivers' interactive behaviors, making it challenging for autonomous vehicles to dynamically assess and respond to other agents during interactions. To address this issue, we propose a framework for assessing drivers' interaction capabilities, oriented towards the interactive process itself, which includes three components: Interaction Risk Perception, Interaction Process Modeling, and Interaction Ability Scoring. We quantify interaction risks through motion state estimation and risk field theory, followed by introducing a dynamic action assessment benchmark based on a game-theoretical rational agent model, and designing a capability scoring metric based on morphological similarity distance. By calculating real-time differences between a driver's actions and the assessment benchmark, the driver's interaction capabilities are scored dynamically. We validated our framework at unsignalized intersections as a typical scenario. Validation analysis on driver behavior datasets from China and the USA shows that our framework effectively distinguishes and evaluates conservative and aggressive driving states during interactions, demonstrating good adaptability and effectiveness in various regional settings.

Tuning-Free Adaptive Style Incorporation for Structure-Consistent Text-Driven Style Transfer

In this work, we target the task of text-driven style transfer in the context of text-to-image (T2I) diffusion models. The main challenge is consistent structure preservation while enabling effective style transfer effects. The past approaches in this field directly concatenate the content and style prompts for a prompt-level style injection, leading to unavoidable structure distortions. In this work, we propose a novel solution to the text-driven style transfer task, namely, Adaptive Style Incorporation~(ASI), to achieve fine-grained feature-level style incorporation. It consists of the Siamese Cross-Attention~(SiCA) to decouple the single-track cross-attention to a dual-track structure to obtain separate content and style features, and the Adaptive Content-Style Blending (AdaBlending) module to couple the content and style information from a structure-consistent manner. Experimentally, our method exhibits much better performance in both structure preservation and stylized effects.