Exploring the new world of retrieval-augmented ML — Generative AI sparked several “wow” moments in 2022. From generative art tools like OpenAI’s DALL-E 2, Midjourney, and Stable Diffusion, to the next generation of Large Language Models like OpenAI’s GPT-3.5 generation models, BLOOM, and chatbots like LaMDA and ChatGPT. It’s hardly surprising that Generative AI is experiencing a boom…

Next-Gen Sentence Embeddings with Multiple Negatives Ranking Loss — Transformer-produced sentence embeddings have come a long way in a very short time. Starting with the slow but accurate similarity prediction of BERT cross-encoders, the world of sentence embeddings was ignited with the introduction of SBERT in 2019 [1]. Since then, many more sentence transformers have been introduced. These models…

How the original sentence transformer (SBERT) was built — Search is entering a golden age. Thanks to “sentence embeddings” and specially trained models called“sentence transformers” we can now search for information using concepts rather than keyword matching. Unlocking a human-like information discovery process. This article will explore the training process of the first sentence transformer, sentence-BERT — more commonly…

How modern language models capture meaning — Transformers have wholly rebuilt the landscape of natural language processing (NLP). Before transformers, we had okay translation and language classification thanks to recurrent neural nets (RNNs) — their language comprehension was limited and led to many minor mistakes, and coherence over larger chunks of text was practically impossible. Since the…

The technology behind Google, Netflix, Spotify, and many more explained — There is perhaps no greater contributor to the success of modern Natural Language Processing (NLP) technology than vector representations of language. The meteoric rise of NLP was ignited with the introduction of word2vec in 2013 [1]. Word2vec is one of the most iconic and earliest examples of dense vectors representing…

How do we organize the world’s most unorganizable data? — Organizing mass amounts of unstructured text data is complicated. It is not designed to be understood by machines, and having humans process this abundance of data is wildly expensive and very slow. Fortunately, there is light at the end of the tunnel. More and more of this unstructured text is…

Unlocking the treasure trove of unstructured data with deep learning — Search is a crucial functionality in many applications and companies globally. Whether in manufacturing, finance, healthcare, or almost any other industry, organizations have vast internal information and document repositories. Unfortunately, the scale of many companies’ data means that the organization and accessibility of information can become incredibly inefficient. …

How to use OpenAI’s Whisper for better speech-enabled (audio) search — OpenAI’s Whisper is a new state-of-the-art (SotA) model in speech-to-text. It can almost flawlessly transcribe speech across dozens of languages and even handle poor audio quality or excessive background noise. The domain of spoken word has always been somewhat out of reach for ML use cases. Whisper changes that for…

Enhancing GIF discovery using NLP and semantic search — Vector search powers some of the most popular services in the world. It serves your Google results, delivers the best podcasts on Spotify, and accounts for at least 35% of consumer purchases on Amazon [1][2]. In this article, we will use vector search applied to language, called semantic search, to…