The 'Travaux d'Initiation à la Recherche' (TIR) is a fundamental component of the Computer Science curriculum designed to immerse students in the scientific approach and critical thinking essential in research. This module is not intended to make you a researcher but to provide a glimpse into what computer science research entails.
Work: Introduction to scientific work on a specific topic. Initiation: A stepping stone into the world of computer science research. Research: Encouragement of critical thinking and continuous skill development.
Common to all CS specialities, addressing research generically. Focus on enhancing critical thinking, an essential skill in professional careers. Teamwork under the guidance of a researcher or client. Completion across three main stages: Literature Search, Full State of the Art, and Setting the Scene.
Our group's focus is on exploring the intersection of Pre-trained Language Models (PLMs), Large Language Models (LLMs), and their application in Text-to-SQL contexts. We aim to investigate various datasets and metrics relevant to this field.
We commence our exploration with a set of key research papers. These are housed in our docs subfolder for easy reference:
Improving Text-to-SQL Evaluation Methodology
KaggleDBQA: Realistic Evaluation of Text-to-SQL Parsers
Objective: To develop a comprehensive understanding of the current research landscape. Output: Bibliography files documenting relevant literature.
Objective: To delve deeper into the literature and develop a nuanced understanding. Output: A detailed written document outlining our findings.
Objective: To synthesize our learning and present our insights. Output: Peer-reviewed content and a poster presentation.