NOSSDAV is a workshop with a long-standing tradition focusing on both emerging and established multimedia systems research topics especially high-risk/high-return ideas. The single-track format brings together academic and industry researchers and practitioners to engage in fruitful and constructive discussions. NOSSDAV encourages papers identifying fundamental open problems in multimedia systems research, advocate new radical ideas or offer counterarguments or critique on the existing state of the art; NOSSDAV prefers exciting-even-if-incomplete papers or reports on promising ideas that may be supported by well-reasoned arguments. Preliminary work can be supported by experimental evidence, analytical insights or well-reasoned technical arguments, without requiring a complete implementation.

NOSSDAV takes a broad perspective on all areas of multimedia systems. Authors are especially encouraged to submit papers with real-world experimental results and datasets. This includes new ideas relating to (but not limited to):

- Network and OS support for AI-generated and AI-serving multimedia systems
- System support for multimodal foundation models and AI-native multimedia
- Agent communication and coordination for multimedia generation and streaming
- Semantic communication and task-oriented multimedia delivery
- Real-time multimedia transport and low-latency communication
- Virtual reality (VR), augmented reality (AR) and immersive systems
- Deep learning approaches to improve multimedia delivery and processing
- Cloud architectures for multimedia coding and processing
- Emerging network systems for multimedia applications
- Medical, surveillance and autonomous driving multimedia systems
- Network-distributed media processing
- Networked GPUs/TPUs, graphics and virtual environments
- Audio, image and video analytics systems
- System support for multimedia interaction with humans
- Internet-scale multimedia systems

Prospective authors are invited to submit an electronic version of full papers, in PDF format, up to six (6) printed pages plus one (1) page solely for references in length (double column ACM style format, https://www.acm.org/publications/proceedings-template) at the workshop Web site. Please use \documentclass[sigconf]{acmart} for preparing your paper. Authors must prepare their papers in a way that preserves the anonymity of the authors (double blind). Please do NOT include the author names under the title. The workshop proceedings will be published by ACM Digital Library. We follow the ACM Policy on Authorship (updated by the ACM Publications Board on May 14, 2026, https://www.acm.org/publications/policies/new-acm-policy-on-authorship). Specifically, be advised of the policy on the use of AI.

Submit by: January 11th, 2027
Notifications on: February 1, 2027
Camera ready on: February 19, 2027
Online Submission: https://nossdav27.hotcrp.com/