This is the landing page for ELFMo’s open-source code, open-access publications, and open data contributions. Unless otherwise specified, content is licensed under the MIT or CC-BY licenses.
Ship-agent is an open-source repository showcasing a maritime data use case developed as part of the ELFMo project: https://github.com/orgs/Ship-agent
Data-augmenter is an open-source repository demonstrating dataset augmentation techniques using large language models, developed by CIC as part of the ELFMo project: https://github.com/CIC-SL/python-data-augmenter
The code repository for the paper titled "Confidence-based Estimators for Predictive Performance in Model Monitoring" published in JAIR in 2025: https://github.com/JuhaniK/AC_trials
The code repository for the paper titled "Performance Estimation in Binary Classification Using Calibrated Confidence" to be published in ACML / Machine Learning journal 2025: https://github.com/JuhaniK/CBPE-experiments
The code repository for the paper titled "Estimating Model Performance Under Covariate Shift Without Labels" to be published in NeurIPS 2025: https://github.com/pape-research/pape_r
The code repository for the Master's thesis "Autonomous LLM Evaluation and Remediation Framework": https://github.com/afnan1456/llm_monitoring_framework/ .
The code repository for the Master's thesis "A Cost-Efficiency Based Framework for Selecting Large Language Models Using Public Benchmark Data" https://github.com/phuvio/total-cost-of-ownership .
The code repository for the Master's thesis "Beyond Generic Responses: Achieving Brand Consistency in AI Chatbots Through Context-Aware Tone Adaptation": https://github.com/Abdulmalik740/Brand_consistent_chatbot
University of Helsinki ELFMo-related publications https://researchportal.helsinki.fi/fi/projects/engineering-large-foundational-models-for-enterprise-integration/publications/
ELFMo is part of the ITEA4 Eureka Cluster on software innovation, supported by funding from local authorities, i.e., ANI, Business Finland, CDTI and VLAIO.
