Enhancing Software Related Information Extraction with Generative Language Models through Single-Choice Question Answering

arXiv:2404.05587v1 Announce Type: new
Abstract: This paper describes our participation in the Shared Task on Software Mentions Disambiguation (SOMD), with a focus on improving relation extraction in scholarly texts through Generative Language Models (GLMs) using single-choice question-answering. The methodology prioritises the use of in-context learning capabilities of GLMs to extract software-related entities and their descriptive attributes, such as distributive information. Our approach uses Retrieval-Augmented Generation (RAG) techniques and GLMs for Named Entity Recognition (NER) and Attributive NER to identify relationships …

abstract arxiv capabilities context cs.cl extraction focus generative improving in-context learning information information extraction language language models methodology paper question question answering software through type