Archive
Blog - posts for September 2021
Sep 27 2021
Facebook GSLM textless NLP
Facebook recently introduced a generative spoken language model (GSLM) called textless NLP.
The research team believes that their GSLM can be an effective method for pre-training downstream tasks trained with few available labelled or annotated data, like spoken summarization, information retrieval tasks, and sentiment analysis.
GSLM uses the latest breakthroughs in representation learning, allowing it to work directly from raw audio signals, without any text or labels. According to Facebook, this opens the door to a new era of textless NLP applications for potentially every language spoken on Earth — even those without significant or limited text datasets. In addition, it enables the development of NLP models that incorporate the full range of expressivity of oral language.
More
Sep 22 2021
SWForum Project Spotlight
SWforumEU Project Spotlight is highlighting DECODER.
Date: 22 September 2021
Place: online
Link to SWForum Project Spotlight
DECODER aims to improve the efficiency of software development & maintenance, and the quality of software in medium criticality systems, such as in IoT, cloud computing and HPC.
Sep 21 2021
Armand Puccetti, CEA List Research Engineer
Developers have to drive through the complexity of software source code
The impact on business is to let Machine Learning help developers and maintainers to produce better code that is easier (and cheaper) to maintain as it is delivered with its related artefacts. After all, project is the result of the transformation of knowledge into code.
Sep 16 2021
NLP-Based Source Code Analysis Tools
Tree Technology, a partner in DECODER Project, posted an article about recent R&D efforts.
Abstract: We have used Natural Language Processing (NLP) techniques in tools aimed to support and improve the software development and software quality processes for Java and C/C++ languages.
The use of complex models has increased performance in many common NLP tasks, such as named entity recognition, text classification, summarisation and translation among others. Besides, transfer learning has also become an interesting option when not much labelled data is available and knowledge learnt from one problem can be applied to a new but related task. In this context, our two NLP-based source code analysis tools - namely Variable Misuse and Code Summarisation - have been conceived by and for software developers.