Awarded Projects

This proposal aims to address the challenges of data scarcity for domain-specific fine-tuning of Large Language Models (LLMs) in languages other than English.

Skeleton-based forensic human identification strongly relies on manual, error-prone methods that can benefit from data-driven automated software alternatives. For this project, experts in generative AI for forensics will contribute to development of a Craniofacial Reconstruction tool.

This project proposes a novel, fully automated dual-loop computational workflow for inverse materials discovery that couples a generative AI model for crystal structure creation with a mixed machine-learned interatomic potential (MLIP) / ab initio (DFT) high-throughput active-learning labeling loop.

Protein design has shifted in the last 2 years by the development of novel generative models for protein design.

Utilizing a robust retrieval-augmented generation (RAG) system enriched with over 150,000 documents, Geo-Llama bridges the limitations of general LLMs in understanding geospatial relationships and OSM's OQL programming language.

Fossil fuels energy production is steeply increasing with correspondingly increasing atmospheric carbon dioxide (CO2) emissions.

This project capitalises on previous AI research utilising LLMs to analyse vast and diverse phage and bacterial genomic datasets to train the first generative model able to adapt phage genomes to new hosts.

The project proposes to perform the first global simulations of plasma fluid turbulence in the W7-AS stellarator. The simulations will be performed with GBS, a three-dimensional turbulence code that solves the drift- reduced Braginskii equations in a toroidal domain and with first-principles sheath

Tremendous progress has been made in magnetic confinement fusion research over the past decades, culminating in the recent world record energy output recorded at JET. However, the challenge of achieving a self-sustaining plasma remains.