Posts by Mina Ghashami

[ad_1] Metrics to measure the gap between neural text and human textImage from unsplash.comRecently, large language models have shown tremendous ability in generating human-like texts. There are many metrics to…

[ad_1] Adapt a pre-trained model to a new domain using HuggingFaceImage from unsplashLarge language models (LLMs) like BERT are usually pre-trained on general domain corpora like Wikipedia and BookCorpus. If…

[ad_1] Hierarchical attention is fasterThis article requires you to have knowledge of standard transformers and how they work. If you are a beginner and you’d like to know about transformers,…

[ad_1] Approximate a matrix without having all of its rows13 min read·13 hours agoImage credit: unsplash.comMatrix approximation is a heavily studied sub-field in data mining and machine learning. A large…