This article provides a comprehensive guide on the basics of BERT (Bidirectional Encoder Representations from Transformers) models. It covers the architecture, use cases, and practical implementations, helping readers understand how to leverage BERT for natural language processing tasks.
An explanation of the differences between encoder- and decoder-style large language model (LLM) architectures, including their roles in tasks such as classification, text generation, and translation.
A detailed guide on creating a text classification model with Hugging Face's transformer models, including setup, training, and evaluation steps.
David Johnson-Davies explains how to efficiently and accurately smooth characters for small displays, using a routine to programmatically smooth out corners and adding hinting functionality.
Snowflake recently announced the launch of Arctic Embed L 2.0 and Arctic Embed M 2.0, two small and powerful embedding models tailored for multilingual search and retrieval. The models are available in medium and large variants, with the medium model incorporating 305 million parameters and the large variant with 568 million parameters. Both models support context lengths of up to 8,192 tokens. They demonstrate high-quality retrieval across multiple languages and excel in benchmarks like MTEB and CLEF.
The article discusses the evolution of search databases and how vector databases are emerging as a powerful alternative to traditional search engines like Elasticsearch.
BEAL is a deep active learning method that uses Bayesian deep learning with dropout to infer the model’s posterior predictive distribution and introduces an expected confidence-based acquisition function to select uncertain samples. Experiments show that BEAL outperforms other active learning methods, requiring fewer labeled samples for efficient training.
A tutorial on using LLM for text classification, addressing common challenges and providing practical tips to improve accuracy and usability.
This article discusses how traditional machine learning methods, particularly outlier detection, can be used to improve the precision and efficiency of Retrieval-Augmented Generation (RAG) systems by filtering out irrelevant queries before document retrieval.
Replace traditional NLP approaches with prompt engineering and Large Language Models (LLMs) for Jira ticket text classification. A code sample walkthrough.
