Autonomous agents will soon run thousands of enterprise workflows, and only organizations with unified, trusted, context-rich data will prevent chaos and unlock reliable value at scale. AI agents are moving beyond coding assistants and customer service chatbots into the operational core of the enterprise. The ROI is promising, but autonomy without alignment is a recipe for chaos. Business leaders need to lay the essential foundations now. Agents are independently handling end-to-end processes across lead generation, supply chain optimization, customer support, and financial reconciliation. A mid-sized organization could easily run 4,000 agents, each making decisions that affect revenue, compliance, and customer experience.

Human-AI shared control allows human to interact and collaborate with autonomous agents to accomplish control tasks in complex environments. Previous Reinforcement Learning (RL) methods attempted goal-conditioned designs to achieve human-controllable policies at the cost of redesigning the reward function and training paradigm. Inspired by the neuroscience approach to investigate the motor cortex in primates, we develop a simple yet effective frequency-based approach called Policy Dissection to align the intermediate representation of the learned neural controller with the kinematic attributes of the agent behavior. Without modifying the neural controller or retraining the model, the proposed approach can convert a given RL-trained policy into a human-controllable policy. We evaluate the proposed approach on many RL tasks such as autonomous driving and locomotion. The experiments show that human-AI shared control system achieved by Policy Dissection in driving task can substantially improve the performance and safety in unseen traffic scenes. With human in the inference loop, the locomotion robots also exhibit versatile controllable motion skills even though they are only trained to move forward. Our results suggest the promising direction of implementing human-AI shared autonomy through interpreting the learned representation of the autonomous agents. Code and demo videos are available at https://metadriverse.github.io/policydissect

The ability of large language models (LLMs) to mimic human-like intelligence has led to a surge in LLM-based autonomous agents. Though recent LLMs seem capable of planning and reasoning given user instructions, their effectiveness in applying these capabilities for autonomous task solving remains underexplored. This is especially true in enterprise settings, where automated agents hold the promise of a high impact. To fill this gap, we propose WorkArena++, a novel benchmark consisting of 682 tasks corresponding to realistic workflows routinely performed by knowledge workers. WorkArena++ is designed to evaluate the planning, problem-solving, logical/arithmetic reasoning, retrieval, and contextual understanding abilities of web agents. Our empirical studies across state-of-the-art LLMs and vision-language models (VLMs), as well as human workers, reveal several challenges for such models to serve as useful assistants in the workplace. In addition to the benchmark, we provide a mechanism to effortlessly generate thousands of ground-truth observation/action traces, which can be used for fine-tuning existing models. Overall, we expect this work to serve as a useful resource to help the community progress towards capable autonomous agents.

In recent years, the remarkable progress of large language models (LLMs) has sparked interest in task automation, which involves decomposing complex tasks described by user instructions into sub-tasks and invoking external tools to execute them, playing a central role in autonomous agents. However, there is a lack of systematic and standardized benchmarks to promote the development of LLMs in task automation. To address this, we introduce TaskBench, a comprehensive framework to evaluate the capability of LLMs in task automation. Specifically, task automation can be divided into three critical stages: task decomposition, tool selection, and parameter prediction. To tackle the complexities inherent in these stages, we introduce the concept of Tool Graph to represent decomposed tasks and adopt a back-instruct method to generate high-quality user instructions.

With the adoption of autonomous vehicles on our roads, we will witness a mixed-autonomy environment where autonomous and human-driven vehicles must learn to coexist by sharing the same road infrastructure. T o attain socially-desirable behaviors, autonomous vehicles must be instructed to consider the utility of other vehicles around them in their decision-making process. Particularly, we study the maneuver planning problem for autonomous vehicles and investigate how a decentralized reward structure can induce altruism in their behavior and incentivize them to account for the interest of other autonomous and human-driven vehicles. This is a challenging problem due to the ambiguity of a human driver's willingness to cooperate with an autonomous vehicle. Thus, in contrast with the existing works which rely on behavior models of human drivers, we take an end-to-end approach and let the autonomous agents to implicitly learn the decision-making process of human drivers only from experience. W e introduce a multi-agent variant of the synchronous Advantage Actor-Critic (A2C) algorithm and train agents that coordinate with each other and can affect the behavior of human drivers to improve traffic flow and safety.Accepted to 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021) W orkshop on Autonomous Driving: Perception, Prediction and Planning

As Reinforcement Learning (RL) agents are increasingly deployed in real-world applications, ensuring their behavior is transparent and trustworthy is paramount. A key component of trust is explainability, yet much of the work in Explainable RL (XRL) focuses on local, single-step decisions. This paper addresses the critical need for explaining an agent's long-term behavior through trajectory-level analysis. We introduce a novel framework that ranks entire trajectories by defining and aggregating a new state-importance metric. This metric combines the classic Q-value difference with a "radical term" that captures the agent's affinity to reach its goal, providing a more nuanced measure of state criticality. We demonstrate that our method successfully identifies optimal trajectories from a heterogeneous collection of agent experiences. Furthermore, by generating counterfactual rollouts from critical states within these trajectories, we show that the agent's chosen path is robustly superior to alternatives, thereby providing a powerful "Why this, and not that?" explanation. Our experiments in standard OpenAI Gym environments validate that our proposed importance metric is more effective at identifying optimal behaviors compared to classic approaches, offering a significant step towards trustworthy autonomous systems.

In this paper, we first observe that policies learned using InRL can overfit to the other agents' policies during training, failing to sufficiently generalize during execution. We introduce a new metric, joint-policy correlation, to quantify this effect.

In this paper we introduce and discuss an approach for multi-agent-oriented visual programming. This aims at enabling individuals without programming experience but with knowledge in specific target domains to design and (re)configure autonomous software. We argue that, compared to procedural programming, it should be simpler for users to create programs when agent abstractions are employed. The underlying rationale is that these abstractions, and specifically the belief-desire-intention architecture that is aligned with human practical reasoning, match more closely with people's everyday experience in interacting with other agents and artifacts in the real world. On top of this, we designed and implemented a visual programming system for agents that hides the technicalities of agent-oriented programming using a blocks-based visual development environment that is built on the JaCaMo platform. To further validate the proposed solution, we integrate the Web of Things (WoT) to let users create autonomous behaviour on top of physical mashups of devices, following the trends in industrial end-user programming. Finally, we report on a pilot user study where we verified that novice users are indeed able to make use of this development environment to create multi-agent systems to solve simple automation tasks.

Among the research topics in multi-agent learning, mixed-motive cooperation is one of the most prominent challenges, primarily due to the mismatch between individual and collective goals.