Chain-of-Thought (CoT) is an efficient prompting method that enables the reasoning ability
of large language models by augmenting the query using multiple examples with …

Unveiling the reasons behind the exceptional success of transformers requires a better
understanding of why attention layers are suitable for NLP tasks. In particular, such tasks …

Understanding the training dynamics of transformers is important to explain the impressive
capabilities behind large language models. In this work, we study the dynamics of training a …

Historically, we are taught to use task-dependent architecture design and objectives to
tackle data science tasks. Counter intuitively, this dogma has been proven (partly) wrong by …

The machine learning and data science community has made significant while dispersive
progress in accelerating transformer-based large language models (LLMs), and one …

While the Transformer architecture has achieved remarkable success across various
domains, a thorough theoretical foundation explaining its optimization dynamics is yet to be …

Questa trattazione si propone di esplorare tecniche avanzate di Machine Learning e Deep
Learning applicate alla previsione dei brillamenti solari. Questi eventi, detti solar flares, sono …

Transformers have become ubiquitous in modern machine learning applications, yet their
training remains a challenging task often requiring extensive trial and error. Unlike previous …

The attention structure is essential for the success of large language models nowadays.
Many works are devoted to a better understanding of attention. In this report, we focus on the …