Blogs on BERT » Lamarr Institute

00 Blog Toborek Vanessa Forschung Sentence Allignment - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Simplified language for all: How to create a dataset in Simple German

06. December 2023

Machine Learning models are only as diverse as the datasets they are trained on. Learn more about the first publicly available bilingual dataset in German and Simple German language provided by us....

00 Blog Paass Video Generation mit Sprachmodellen - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Basics

From word to video: How foundation models can be used to create moving images from text

25. January 2023

Large language models have immense transformation potential. So-called foundation models can now not only capture and generate the meaning of texts and images, but can also be applied to videos....

00 Blog Sifa ART 9 - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Application

ART – Automated Tool for Generating Financial Statements

16. March 2022

Creating financial reports takes a lot of time and is expensive. ART is a new method that should make it possible to convert structured table data from financial reports into text....

Research

How to Evaluate Chat Quality Using Standard NLP Benchmarks

02. March 2022

Until now, language model-based assessments of chatbots have only produced a score for overall quality, often neglecting the context of the dialogue. With models that are trained on GLUE tasks, this has come to an end....

00 Blog Nedelchev Smalltalk - Lamarr Institute for Machine Learning (ML) and Artificial Intelligence (AI)

Research

Evaluating small talk using language models: Language model transformers as evaluators for open-domain dialogues

23. February 2022

Dialogsysteme sind in Sprachassistenten wie Siri oder Alexa oder in Chatbots auf Social-Media-Plattformen weit verbreitet. Aber woher weiß man, dass der eigens entwickelte Chatbot tatsächlich funktioniert?...

Research

BERT: How Do Vectors Accurately Describe the Meaning of Words?

12. March 2021

BERT makes it possible to solve many problems of natural language interpretation. This article explains in detail the architecture, properties and functionality of the model developed by Google....

Basics

Capturing the meaning of words through vectors

10. March 2021

The meaning of words can be represented by numerical vectors that also include the neighboring words. This allows syntactic and semantic relationships of the language to be captured and used to interpret meanings....