CarePilot: A Multi-Agent Framework for Long-Horizon Computer Task Automation in Healthcare

• URL: http://arxiv.org/abs/2603.24157v1
• Date: Wed, 25 Mar 2026 10:25:48 GMT
• Title: CarePilot: A Multi-Agent Framework for Long-Horizon Computer Task Automation in Healthcare
• Abstract summary: Multimodal agentic pipelines are transforming human-computer interaction by enabling efficient and accessible automation of complex, real-world tasks.<n>Recent efforts have focused on short-horizon or general-purpose applications, leaving long-horizon automation for domain-specific systems, particularly in healthcare, largely unexplored.<n>We introduce CarePilot, a multi-agent framework based on the actor-critic paradigm.<n>Our experiments show that CarePilot achieves state-of-the-art performance, outperforming strong closed-source and open-source multimodal baselines by approximately 15.26% and 3.38%, respectively.
• Score: 37.42599407869901
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multimodal agentic pipelines are transforming human-computer interaction by enabling efficient and accessible automation of complex, real-world tasks. However, recent efforts have focused on short-horizon or general-purpose applications (e.g., mobile or desktop interfaces), leaving long-horizon automation for domain-specific systems, particularly in healthcare, largely unexplored. To address this, we introduce CareFlow, a high-quality human-annotated benchmark comprising complex, long-horizon software workflows across medical annotation tools, DICOM viewers, EHR systems, and laboratory information systems. On this benchmark, existing vision-language models (VLMs) perform poorly, struggling with long-horizon reasoning and multi-step interactions in medical contexts. To overcome this, we propose CarePilot, a multi-agent framework based on the actor-critic paradigm. The Actor integrates tool grounding with dual-memory mechanisms (long-term and short-term experience) to predict the next semantic action from the visual interface and system state. The Critic evaluates each action, updates memory based on observed effects, and either executes or provides corrective feedback to refine the workflow. Through iterative agentic simulation, the Actor learns to perform more robust and reasoning-aware predictions during inference. Our experiments show that CarePilot achieves state-of-the-art performance, outperforming strong closed-source and open-source multimodal baselines by approximately 15.26% and 3.38%, respectively, on our benchmark and out-of-distribution dataset.

Related papers

• GTA: Generating Long-Horizon Tasks for Web Agents at Scale [82.43869456830664]
  We introduce a scalable framework, GTA, that integrates crawling, retrieval-based seeding, in-context generation, and automated quality control.<n>We instantiate the pipeline on over 50 websites covering e-commerce, government, forums, and news, with multilingual and multi-hop coverage.<n>Our contributions are three-fold: (i) formalizing multi-hop web-agent task generation, (ii) proposing an efficient and validated pipeline for automatic data creation, and (iii) releasing a dynamic benchmark with reproducible evaluation.
  arXiv  Detail & Related papers  (2026-05-28T01:05:50Z)
• Exploring Interaction Paradigms for LLM Agents in Scientific Visualization [7.22431217973039]
  This paper examines how different types of large language model (LLM) agents perform on scientific visualization (SciVis) tasks.<n>We compare three primary interaction paradigms, including domain-specific agents with structured tool use, computer-use agents, and general-purpose coding agents.
  arXiv  Detail & Related papers  (2026-04-30T15:22:28Z)
• RoboAgent: Chaining Basic Capabilities for Embodied Task Planning [46.248451288196435]
  This paper focuses on embodied task planning, where an agent acquires visual observations from the environment and executes atomic actions to accomplish a given task.<n>We propose RoboAgent, a capability-driven planning pipeline in which the model actively invokes different sub-capabilities.<n>We use a multi-stage paradigm that consists of: (1) behavior cloning with expert plans, (2) DAgger training using trajectories collected by the model, and (3) reinforcement learning guided by an expert policy.
  arXiv  Detail & Related papers  (2026-04-09T04:01:27Z)
• MedSPOT: A Workflow-Aware Sequential Grounding Benchmark for Clinical GUI [0.7552557021953206]
  We introduce MedSPOT, a workflow-aware sequential grounding benchmark for clinical GUI environments.<n> benchmark comprises 216 task-driven videos with 597 annotateds, in which each task consists of 2 to 3 interdependent grounding steps.<n>We also introduce a comprehensive failure taxonomy, including edge bias, small-target errors, no prediction, near miss, far miss, and toolbar confusion.
  arXiv  Detail & Related papers  (2026-03-20T14:43:53Z)
• MindWatcher: Toward Smarter Multimodal Tool-Integrated Reasoning [55.221850286246]
  We introduce MindWatcher, a tool-integrated reasoning agent with interleaved thinking and multimodal chain-of-thought (CoT) reasoning.<n>MindWatcher can autonomously decide whether and how to invoke diverse tools and coordinate their use.<n>A large-scale, high-quality local image retrieval database, covering eight categories including cars, animals, and plants, endows model with robust object recognition.
  arXiv  Detail & Related papers  (2025-12-29T12:16:12Z)
• ReX-MLE: The Autonomous Agent Benchmark for Medical Imaging Challenges [5.886200278450183]
  We introduce ReX-MLE, a benchmark of 20 challenges derived from high-impact medical imaging competitions.<n>Unlike prior benchmarks, ReX-MLE evaluates full end-to-end, requiring agents to independently manage data preprocessing, model training, and submission.<n>We observe a severe performance gap: most submissions rank in the 0th percentile compared to human experts.
  arXiv  Detail & Related papers  (2025-12-19T17:44:40Z)
• LoCoBench-Agent: An Interactive Benchmark for LLM Agents in Long-Context Software Engineering [90.84806758077536]
  We introduce textbfLoCoBench-Agent, a comprehensive evaluation framework specifically designed to assess large language models (LLMs) agents in realistic, long-context software engineering.<n>Our framework extends LoCoBench's 8,000 scenarios into interactive agent environments, enabling systematic evaluation of multi-turn conversations.<n>Our framework provides agents with 8 specialized tools (file operations, search, code analysis) and evaluates them across context lengths ranging from 10K to 1M tokens.
  arXiv  Detail & Related papers  (2025-11-17T23:57:24Z)
• Spider2-V: How Far Are Multimodal Agents From Automating Data Science and Engineering Workflows? [73.81908518992161]
  We introduce Spider2-V, the first multimodal agent benchmark focusing on professional data science and engineering. Spider2-V features real-world tasks in authentic computer environments and incorporating 20 enterprise-level professional applications. These tasks evaluate the ability of a multimodal agent to perform data-related tasks by writing code and managing the GUI in enterprise data software systems.
  arXiv  Detail & Related papers  (2024-07-15T17:54:37Z)
• WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks? [83.19032025950986]
  We study the use of large language model-based agents for interacting with software via web browsers. WorkArena is a benchmark of 33 tasks based on the widely-used ServiceNow platform. BrowserGym is an environment for the design and evaluation of such agents.
  arXiv  Detail & Related papers  (2024-03-12T14:58:45Z)

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.