The Pervasive Agentic Systems and Multimodal Foundation Models (PerAgent) workshop focuses on the critical systems-level challenges of transitioning from narrow, singletask edge AI to general-purpose, autonomous AI agents operating in physical environments. Currently, foundation models are trapped in the cloud due to massive computational requirements. This workshop explores how large-scale models such as Large Language Models (LLMs), Vision-Language Models (VLMs), and Multimodal Language Models (MLMs) can be compressed, decentralized, and securely integrated with pervasive sensor networks.

Topics of interest include but are not limited to:

*Hardware-software co-design for edge agents
*Decentralized and split/federated agentic learning
*Multimodal sensor fusion for VLMs/MLMs
*Trust, safety, and local overrides for physical actuation
*Bandwidth-efficient distributed inference
*Cross-domain agent adaptation algorithms
*Agentic synchronization across IoT mesh networks



Paper Requirements
Workshop papers follow the standard PerCom workshop length limit: 6 pages.
Authors may include 1 additional page, for a maximum of 7 pages total (including references).
Papers should contain names and affiliations of the authors (not blinded).
All papers must be typeset in double-column IEEE format using 10pt fonts on US letter paper, with all fonts embedded.
All accepted papers must be presented in person by a designated presenter, preferably one of the authors.
Proxy or remote presentations are not allowed unless pre-approved by the workshop chairs, general chairs, and steering committee at least one week in advance.

Use the official IEEE LaTeX or Microsoft Word conference templates and follow the two-column formatting instructions.