GreenMM: Efficient Systems and AI for the Planet

Venue: Rio de Janeiro, Brazil, In conjunction with the 34th ACM International Conference on Multimedia.

Introduction

GreenMM brings together researchers working on efficient multimedia AI systems and multimedia AI for environmental and planetary applications. The workshop focuses on how to build multimedia models that are both technically strong and environmentally responsible. Multimedia AI spans vision, video, audio, language, and multimodal learning. It is now one of the most computationally demanding areas of AI, while also becoming central to biodiversity monitoring, climate analysis, environmental sensing, agriculture, and disaster response.

Topics

GreenMM welcomes methodological, systems, and application-driven work across sustainable AI and environmental multimedia research.

Efficient and Sustainable Multimedia AI Systems

Energy- and memory-efficient video and multimodal architectures
Efficient vision-language and multimodal foundation models
Compression, pruning, quantization, and distillation for multimedia
Sparse, adaptive, and token-reduction methods for vision and video
Parameter-efficient fine-tuning for large multimodal models
Efficient generative models for image, audio, and video
Carbon-aware training, scheduling, and deployment
Resource-efficient inference for edge and real-time systems
Data-efficient learning under limited annotations
Benchmarking and reporting of energy and carbon footprint

Multimedia AI for Environmental and Planetary Applications

Vision and video analysis for biodiversity and wildlife monitoring
Acoustic and audio-based monitoring of ecosystems and species
Satellite and aerial imagery analysis for land-use and climate change
Environmental change detection in long-term multimodal streams
AI for precision agriculture and sustainable food systems
Multimedia sensing for water, air, and soil quality
Disaster monitoring and climate-risk analysis
Vision-language models for environmental reporting
Low-cost, scalable sensing systems for environmental monitoring
Integration of video, audio, satellite, and text data

Format

Prospective authors are invited to submit high-quality full papers representing original results. Submissions will follow the ACM MM 2026 format. Note that paper submissions must adhere to the main conference submission style, format, length restrictions, and camera-ready timeline. Submission link: Soon.

Presentation policy: ACM Multimedia 2026 is an on-site event only. This means that all papers and contributions must be presented by a physical person on-site; remote presentations will not be hosted or allowed. Papers and contributions not presented on-site will be considered a no-show and removed from the proceedings of the conference. More details will be provided to handle unfortunate situations in which none of the authors would be able to attend the conference physically.