Call For Papers

The GREEN-AI workshop invites submissions addressing methodological, empirical, and theoretical advances toward environmentally sustainable artificial intelligence. As AI systems continue to scale in size and computational demands, understanding and reducing their environmental footprint has become a central research challenge. GreenAI focuses on the sustainability of AI itself, spanning algorithm design, model development, evaluation, infrastructure efficiency, and lifecycle-aware deployment. The workshop aims to foster discussion and innovation around integrating environmental considerations as first-class objectives in machine learning research.

The GREEN-AI workshop invites submissions addressing methodological, empirical, and theoretical advances toward environmentally sustainable artificial intelligence. As AI systems continue to scale in size and computational demands, understanding and reducing their environmental footprint has become a central research challenge. GreenAI focuses on the sustainability of AI itself, spanning algorithm design, model development, evaluation, infrastructure efficiency, and lifecycle-aware deployment. The workshop aims to foster discussion and innovation around integrating environmental considerations as first-class objectives in machine learning research.

We welcome contributions including, but not limited to:

• Energy-Efficient Algorithms: Techniques for reducing computational cost and energy consumption in training and inference of AI models, including large language models (LLMs), multimodal systems, agentic systems and retrieval-augmented generation (RAG) frameworks.

• Resource-Aware AI Systems: Design and deployment of AI architectures that optimize hardware utilization, cloud and edge resource efficiency, and overall carbon footprint reduction.

• Evaluation Metrics and Benchmarks: Development of metrics and benchmarks that integrate energy efficiency, sustainability, and carbon impact alongside traditional AI performance measures.

• Sustainable Data Practices: Approaches to dataset creation, maintenance, and curation that minimize environmental costs, including strategies for data reuse, deduplication, and responsible scaling.

• Industry–Academia Collaboration: Case studies, lessons learned, and best practices from cross-sector efforts to design, deploy, and monitor environmentally responsible AI systems.

• Societal and Cross-Domain Applications: Exploration of sustainable AI practices across domains such as health, media, transportation, agriculture, and environmental monitoring.

We invite three types of submissions:

• Full Papers (up to 12 pages, excluding references): Mature and original research contributions with comprehensive experimental or theoretical validation.

• Short Papers (up to 6 pages, excluding references): Preliminary findings, negative results, methodological insights, novel research ideas, position papers, or early-stage work intended to stimulate discussion and open new research directions.

• Extended Abstract (2–4 pages, excluding references - non archival): Work in progress, previously published research seeking discussion, or position statements relevant to sustainable and resource-aware AI.

All submissions must follow the CEUR formatting guidelines and must be anonymized for double-blind review.

We plan to publish accepted papers in CEUR Workshop Proceedings, subject to receiving a sufficient number of submissions.