We derive a minimax distributionally robust inverse reinforcement learning (IRL) algorithm to reconstruct the utility functions of a multi-agent sensing system. Specifically, we construct utility estimators which minimize the worst-case prediction error over a Wasserstein ambiguity set centered at noisy signal observations. We prove the equivalence between this robust estimation and a semi-infinite optimization reformulation, and we propose a consistent algorithm to compute solutions. We illustrate the efficacy of this robust IRL scheme in numerical studies to reconstruct the utility functions of a cognitive radar network from observed tracking signals.

Multi-Agent Path Finding (MAPF) is the problem of effectively finding efficient collision-free paths for a group of agents in a shared workspace. The MAPF community has largely focused on developing high-performance heuristic search methods. Recently, several works have applied various machine learning (ML) techniques to solve MAPF, usually involving sophisticated architectures, reinforcement learning techniques, and set-ups, but none using large amounts of high-quality supervised data. Our initial objective in this work was to show how simple large scale imitation learning of high-quality heuristic search methods can lead to state-of-the-art ML MAPF performance. However, we find that, at least with our model architecture, simple large scale (700k examples with hundreds of agents per example) imitation learning does \textit{not} produce impressive results. Instead, we find that by using prior work that post-processes MAPF model predictions to resolve 1-step collisions (CS-PIBT), we can train a simple ML MAPF model in minutes that dramatically outperforms existing ML MAPF policies. This has serious implications for all future ML MAPF policies (with local communication) which currently struggle to scale. In particular, this finding implies that future learnt policies should (1) always use smart 1-step collision shields (e.g. CS-PIBT), (2) always include the collision shield with greedy actions as a baseline (e.g. PIBT) and (3) motivates future models to focus on longer horizon / more complex planning as 1-step collisions can be efficiently resolved.

Despite advancements in perception and planning for autonomous vehicles (AVs), validating their performance remains a significant challenge. The deployment of planning algorithms in real-world environments is often ineffective due to discrepancies between simulations and real traffic conditions. Evaluating AVs planning algorithms in simulation typically involves replaying driving logs from recorded real-world traffic. However, agents replayed from offline data are not reactive, lack the ability to respond to arbitrary AV behavior, and cannot behave in an adversarial manner to test certain properties of the driving policy. Therefore, simulation with realistic and potentially adversarial agents represents a critical task for AV planning software validation. In this work, we aim to review current research efforts in the field of adversarial and reactive traffic agents, with a particular focus on the application of classical and adversarial learning-based techniques. The objective of this work is to categorize existing approaches based on the proposed scenario controllability, defined by the number of reactive or adversarial agents. Moreover, we examine existing traffic simulations with respect to their employed default traffic agents and potential extensions, collate datasets that provide initial driving data, and collect relevant evaluation metrics.

The rapid evolution of Multimodal Large Language Models (MLLMs) has brought substantial advancements in artificial intelligence, significantly enhancing the capability to understand and generate multimodal content. While prior studies have largely concentrated on model architectures and training methodologies, a thorough analysis of the benchmarks used for evaluating these models remains underexplored. This survey addresses this gap by systematically reviewing 211 benchmarks that assess MLLMs across four core domains: understanding, reasoning, generation, and application. We provide a detailed analysis of task designs, evaluation metrics, and dataset constructions, across diverse modalities. We hope that this survey will contribute to the ongoing advancement of MLLM research by offering a comprehensive overview of benchmarking practices and identifying promising directions for future work. An associated GitHub repository collecting the latest papers is available.

The rapid advancement in generative pre-training models is propelling a paradigm shift in technological progression from basic applications such as chatbots towards more sophisticated agent-based systems. It is with huge potential and necessity that the 6G system be combined with the copilot of large language model (LLM) agents and digital twins (DT) to manage the highly complicated communication system with new emerging features such as native AI service and sensing. With the 6G-oriented agent, the base station could understand the transmission requirements of various dynamic upper-layer tasks, automatically orchestrate the optimal system workflow. Through continuously get feedback from the 6G DT for reinforcement, the agents can finally raise the performance of practical system accordingly. Differing from existing LLM agents designed for general application, the 6G-oriented agent aims to make highly rigorous and precise planning with a vast amount of extra expert knowledge, which inevitably requires a specific system design from model training to implementation. This paper proposes a novel comprehensive approach for building task-oriented 6G LLM agents. We first propose a two-stage continual pre-training and fine-tuning scheme to build the field basic model and diversities of specialized expert models for meeting the requirements of various application scenarios. Further, a novel inference framework based on semantic retrieval for leveraging the existing communication-related functions is proposed. Experiment results of exemplary tasks, such as physical-layer task decomposition, show the proposed paradigm's feasibility and effectiveness.

In recent years, Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse NLP tasks. Extensive research has explored how to enhance the logical reasoning abilities such as Chain-of-Thought, Chain-of-Thought with Self-Consistency, Tree-Of-Thoughts, and multi-agent debates. In the context of multi-agent debates, significant performance improvements can be achieved with an increasing number of agents and debate rounds. However, the escalation in the number of agents and debate rounds can drastically raise the tokens cost of debates, thereby limiting the scalability of the multi-agent debate technique. To better harness the advantages of multi-agent debates in logical reasoning tasks, this paper proposes a method to significantly reduce token cost in multi-agent debates. This approach involves dividing all agents into multiple debate groups, with agents engaging in debates within their respective groups and sharing interim debate results between groups. Comparative experiments across multiple datasets have demonstrated that this method can reduce the total tokens by up to 51.7% during debates and while potentially enhancing accuracy by as much as 25%. Our method significantly enhances the performance and efficiency of interactions in the multi-agent debate.

In recent years, the e-commerce industry has seen a rapid increase in the demand for advanced AI-driven customer service solutions. Traditional cloud-based models face limitations in terms of latency, personalized services, and privacy concerns. Furthermore, end devices often lack the computational resources to deploy large AI models effectively. In this paper, we propose an innovative End-Cloud Collaboration (ECC) framework for advanced AI customer service in e-commerce. This framework integrates the advantages of large cloud models and mid/small-sized end models by deeply exploring the generalization potential of cloud models and effectively utilizing the computing power resources of terminal chips, alleviating the strain on computing resources to some extent. Specifically, the large cloud model acts as a teacher, guiding and promoting the learning of the end model, which significantly reduces the end model's reliance on large-scale, high-quality data and thereby addresses the data bottleneck in traditional end model training, offering a new paradigm for the rapid deployment of industry applications. Additionally, we introduce an online evolutive learning strategy that enables the end model to continuously iterate and upgrade based on guidance from the cloud model and real-time user feedback. This strategy ensures that the model can flexibly adapt to the rapid changes in application scenarios while avoiding the uploading of sensitive information by performing local fine-tuning, achieving the dual goals of privacy protection and personalized service. %We make systematic contributions to the customized model fine-tuning methods in the e-commerce domain. To conclude, we implement in-depth corpus collection (e.g., data organization, cleaning, and preprocessing) and train an ECC-based industry-specific model for e-commerce customer service.

While robust optimal control theory provides a rigorous framework to compute robot control policies that are provably safe, it struggles to scale to high-dimensional problems, leading to increased use of deep learning for tractable synthesis of robot safety. Unfortunately, existing neural safety synthesis methods often lack convergence guarantees and solution interpretability. In this paper, we present Minimax Actors Guided by Implicit Critic Stackelberg (MAGICS), a novel adversarial reinforcement learning (RL) algorithm that guarantees local convergence to a minimax equilibrium solution. We then build on this approach to provide local convergence guarantees for a general deep RL-based robot safety synthesis algorithm. Through both simulation studies on OpenAI Gym environments and hardware experiments with a 36-dimensional quadruped robot, we show that MAGICS can yield robust control policies outperforming the state-of-the-art neural safety synthesis methods.

Task allocation in multi-human multi-robot (MH-MR) teams presents significant challenges due to the inherent heterogeneity of team members, the dynamics of task execution, and the information uncertainty of operational states. Existing approaches often fail to address these challenges simultaneously, resulting in suboptimal performance. To tackle this, we propose ATA-HRL, an adaptive task allocation framework using hierarchical reinforcement learning (HRL), which incorporates initial task allocation (ITA) that leverages team heterogeneity and conditional task reallocation in response to dynamic operational states. Additionally, we introduce an auxiliary state representation learning task to manage information uncertainty and enhance task execution. Through an extensive case study in large-scale environmental monitoring tasks, we demonstrate the benefits of our approach.

Imitation learning enables robots to learn new tasks from human examples. One current fundamental limitation while learning from humans is causal confusion. Causal confusion occurs when the robot's observations include both task-relevant and extraneous information: for instance, a robot's camera might see not only the intended goal, but also clutter and changes in lighting within its environment. Because the robot does not know which aspects of its observations are important a priori, it often misinterprets the human's examples and fails to learn the desired task. To address this issue, we highlight that -- while the robot learner may not know what to focus on -- the human teacher does. In this paper we propose that the human proactively marks key parts of their task with small, lightweight beacons. Under our framework the human attaches these beacons to task-relevant objects before providing demonstrations: as the human shows examples of the task, beacons track the position of marked objects. We then harness this offline beacon data to train a task-relevant state embedding. Specifically, we embed the robot's observations to a latent state that is correlated with the measured beacon readings: in practice, this causes the robot to autonomously filter out extraneous observations and make decisions based on features learned from the beacon data. Our simulations and a real robot experiment suggest that this framework for human-placed beacons mitigates causal confusion and enables robots to learn the desired task from fewer demonstrations. See videos here: https://youtu.be/oy85xJvtLSU