NLP

A case study on WiC dataset — Written by Sinan Gultekin and Leonardo Rigutini The race for the leaderboard Nowadays, ever-growing attention to dataset and construction is a natural outcome of examining what kind of behaviours are learnt by models. This attention leads researchers to build benchmark platforms, such as the GLUE and its successor SuperGLUE [1] benchmarks, which aim to become…

In 2020, we have trained and open-sourced the first Dutch GPT2 model, in various sizes. Of course we wanted to share this with the world by… — In 2020, we have trained and open-sourced the first Dutch GPT2 model, in various sizes. Of course we wanted to share this with the world by open-sourcing the models, the code and a nice application that showcases its use. But this nice application comes at a cost, literally…

How Does the Picasso of AI work? — A few days ago, OpenAI released what I find to be the most striking display of the creative power of AI: DALL-E 2. On the most basic level, DALLE-2 is a function that maps text to images with remarkable accuracy, producing high quality and vibrant output images. But how does…

Motivation — Motivation Goodreads.com is possibly the most popular website for book reviewers, I frequently visit it for gauging public reception before reading a book or to find similar ones to my favorite reads. However, popular books have thousands of reviews, so it is usually very difficult and time consuming to learn more…

And How NLG Is Vital To Truly Conversational Agents —

Have you ever wondered if extractive QA models contain answers? Read on to learn more! — When people approach natural language processing, and specifically question answering, for the first time, there are a few concepts that cause confusion around models, training them, and the need for data: Will I need to train a model on my own data? Where do the answers come from? And so…

Challenges of generating language — Introduction Siri, where does Santa live? “He’s in your heart and at The North Pole. Mostly at the North Pole,” comes the tart, prompt reply from Apple’s famous digital assistant. Technologies like Siri rely on machine learning, big data, and natural language processing (NLP) to respond to human requests in as…

All you need to know to build a full demo: Hugging Face Hub, Tensorboard, Streamlit, and Hugging Face Spaces — Hello fellow NLP enthusiasts! I’m back at writing articles after a small break, can’t wait to dig into new NLP topics! In this article, we see a complete example of fine-tuning of T5 for generating candidate titles for articles. The model is fine-tuned entirely on Colab, we visualize its training…