This week on Journal Club session Harpreet Singh will talk about his recent work on "Survey of Sentiment Analysis models leveraging transformers along with other NLP techniques to enrich word vectors for classification".

In recent years, the development of large language models (LLMs) such as the GPT series has revolutionized natural language processing (NLP) research, rendering many specialized models obsolete. Rather than focusing on individual tasks, researchers are now prioritizing the creation of general-purpose language understanding systems capable of performing multiple tasks. Competitive evaluation criteria, such as the Glue score, are being used to evaluate these models. However, due to the limited resources of many universities, they are unable to undertake such research and instead continue to work on text classification tasks. Sentiment analysis is a popular text classification task, and researchers are now incorporating transformers and other techniques to improve performance on sentiment analysis datasets. In this journal club meeting, we will discuss various models that leverage transformers and other techniques to enhance sentiment analysis performance.

Papers:

• TBA.

Date: 2023/03/17
Time: 14:00
Location: online