Skip to main content
The European High Performance Computing Joint Undertaking (EuroHPC JU)

Scaling Laws for Energy-Based Language Models: Towards Natively Explainable Enterprise AI

50 000 Awarded Resources (in node hours)
Leonardo BOOSTER System Partition
February 2026 - 12 months Allocation Period

This project proposes to train and validate a novel language model architecture based on energy-based neural networks that provides native explainability — a critical requirement under the EU AI Act for high-risk AI systems. Unlike standard transformer architectures, our approach produces built-in confidence scores and adaptive compute allocation as natural byproducts of inference, without requiring post-hoc approximation methods.

This project will: 

(1) establish scaling laws for this architecture class from 59M to 7B parameters, 

(2) benchmark against equivalent transformer baselines at each scale, and 

(3) validate the architecture for enterprise deployment in regulated sectors.

Preliminary results on 13M-parameter models demonstrate competitive performance with standard transformers while providing native confidence metrics at zero additional inference cost. EuroHPC JU resources will enable scaling these results to 1B-7B parameters.

Principal Investigator, Company and Country

Marius Dima, Qriton Technologies, Romania