Overview
Agentic computing has become more prominent than ever, driven by advancements in on-device LLMs, interest in cross-application automation, and involving humans in the loop of autonomous systems. AutoSys aims to provide a venue for presenting current research and technology trends and for debating future research agendas in autonomous agents.
We solicit papers of 6 or fewer pages that present preliminary research in broad areas of mobile agents, including prototyping a system, experience in designing a novel technology, or a survey of useful tools for designing interdisciplinary systems and applications. We also encourage position papers that propose new research directions or advocate disruptive design ideas.
Topics of Interest
Topics of interest include (but are not limited to):
- System Architecture for On-device Agents
- Workflow Automation and Tool Orchestration
- Multimodal Sensing for Context-Aware Agents
- Security, Privacy, and Trust in Agent Actions
- Large Language Models (LLMs) for Mobile Tasks
- Human-Agent Interaction (HAI) Systems
- Operating System Primitives for Autonomy
- Energy-Efficient Always-on Reasoning
- Agent Collaboration in IoT Networks
- Real-world Deployment and Case Studies
- Evaluation Benchmarks for Mobile Agents
Submission Guidelines
Authors should submit via HotCRP. We solicit two types of submissions:
Extended abstracts for talks
1–2 pages (excluding references).
Suitable for early-stage ideas, position statements, ongoing projects, demos, and insights from production systems. We strongly welcome contributions from industry practitioners and the open source community to share real-world experiences and challenges. Accepted abstracts will be presented at the workshop.
Short/position papers
Up to 6 pages (excluding references).
Suitable for more complete concepts, research results, experience reports, or comprehensive position papers.
All submissions must follow the ACM double-column conference format. Font size no smaller than 10 points. The review process is single-blind, with each submission receiving at least two reviews.