Welcome

The low-power edge AI for industrial applications workshop aims to bring together researchers and engineers focusing on efficient, edge AI and neuromorphic models for low-power edge platforms. It is co-located with IJCNN and will take place in Maastricht in 24th of June 2026.

Background

The widespread adoption of Artificial Intelligence (AI) in industrial environments, ranging from manufacturing and logistics to predictive maintenance and robotics, is limited by high energy and computational demands of modern neural networks. While cloud-based AI has driven major advances in perception, control and optimization, it remains unsuitable with latency-sensitive, privacy-critical and power-constrained industrial processes. The future of intelligent industry lies in edge AI: enabling low-power, adaptive and trustworthy intelligence directly to embedded and autonomous systems. The Workshop on Low-Power Edge AI for Industrial Applications (LPAI4IA) aims to gather researchers and practitioners at the intersection of efficient AI design, embedded computing, IoT and industrial deployment. It focuses on enabling advanced neural architectures, such as transformers, spiking and graph neural networks, to operate under strict energy and real-time constraints, while maintaining reliability and interpretability. Emphasis is given on cross-layer innovation, including model compression, neuromorphic computing, hardware-aware training and adaptive edge inference. Contributions addressing trustworthy and explainable AI in industrial and safety-critical scenarios are also encouraged.

Call for Papers

The workshop will cover a range of topics related to low-power edge AI, including:

• Efficient and Tiny Deep Neural Networks. Model compression, quantization, pruning, transfer learning and low-rank adaptation for energy efficiency and adaptive inference.

• Neuromorphic Systems and Spiking Neural Networks. Event-driven computation and hardware-software co-design for ultra-low-power sensing and actuation.

• Interpretable and Explainable AI / Trustworthy and Reliable AI. Transparent decision-making for safety-critical industrial operations, with cybersecurity as a particularly relevant application area.

• Federated Learning, Reinforcement Learning, Continual Learning and Cognitive Models. Self-optimizing edge systems capable of learning from evolving industrial environments.

• Neurosymbolic and Physics-inspired Neural Networks. Incorporating domain knowledge and physical constraints into compact and reliable edge models.

• Industrial Applications of Neural Networks. Deployment and validation of low-power AI in manufacturing, predictive maintenance, autonomous robotics, energy systems and process control.

• Graph Neural Networks, Transformer Networks, Mixture of Experts and Large Language Models. Efficient adaptation and distillation of large architectures for resource-constrained industrial devices.

Important dates and format

Papers will be reviewed by the workshop's technical program committee according to criteria regarding a submission's quality, relevance to the workshop's topics and its potential to spark discussions about directions, insights and solutions on the topics mentioned above. Research papers, case studies, position papers and work-in-progress are all welcome.

To foster the broadest possible engagement and exchanging of ideas, LPAI4IA does not claim copyright, making it possible for authors of accepted contributions to present work that has already been published or is in the process of being published elsewhere. With the explicit permission of the authors, accepted papers will be posted on the workshop website prior to the event.

LPAI4IA welcomes two types of contributions: research papers and work-in-progress papers/early ideas:

• Research/Long Papers

Long papers (not exceeding 8 pages in double-column or 15 pages in LNCS format) can be submitted using any of the commonly used templates (e.g., ACM, IEEE, LNCS).

• Short Papers

Submissions of early ideas, work-in-progress papers or negative/failed research results require a short paper of at most 2 pages (not including references) in IEEE double-column format or 4 pages (not including references) in LNCS single-column format.

Submission Start: Mar 15 2026 11:59PM UTC

Submission Deadline: May 09 2026 11:59PM UTC

Notification of results: May 20 2026

Submission site: OpenReview 

Program Committee

Invited speakers

Technical Program Committee

To be announced soon...