News

Recently, some members of the Diverso Lab attended the highly esteemed AAAI'23 conference in Washington DC, which focuses on artificial intelligence. Their visit served two purposes. Firstly, they presented their contribution, "FASTDIAGP: An Algorithm for Parallelized Direct Diagnosis," which resulted from a collaboration with TU Graz and former PhD students from Chile. Secondly, they aimed to gain knowledge and comprehension of the current state of the art in artificial intelligence, particularly in the areas of LLM (Large Language Models) and constraint problem solving. The conference was divided into three main blocks. The first block was the workshops program, which comprised a series of talks and workshops focused on specific techniques in AI. One workshop attended was on program synthesis, which featured various talks on how to automatically synthesize programs, mainly…

natural language processing field connecting the advances with software engineering topics.   The talk addressed the role of scaling a deep learning model to achieve unprecedented results across the whole set of natural language processing evaluations. This exposition of the scaling principles began with the critical property of model parallelism and the attention mechanism, first introduced in 2017 by the famous article "Attention is all you need." Then Antonio exposed the scaling laws, which dictate the role of the model number of parameters, training data, amount of tokens, and compute time are the fundamental building blocks for optimal training. The talk showed the most influential model in the code generation scenario by the hand of DeepMind's Alpha Code. Antonio's latest research addresses mono and cross-lingual code using transformer-based language models.…