Control Ai: Aleph Alpha Introduces New Generation Of Explainable Ai Models

Aleph Alpha, a Heidelberg-based AI start-up, has introduced a new generation of its AI models, the Luminous LLM family. This family consists of three Foundation Models of different sizes and now includes a Control version. This version is technically an Instruct model that has been readjusted in an automated process without human feedback. The new model generation is characterized by explainability and reproducibility of the output. Aleph Alpha has published research contributions together with academic partners in Spring 2023, such as Attention Manipulation and a new method for containing the distribution bias in large language models. The implemented functions are steps towards the upcoming regulatory requirements for explainability and traceability.