Project area/S
- Computer Science/ Astronomy
Project Details
The proposed project seeks to explore integrating large language models with astronomy visualization software to enhance user experience and facilitate interactive exploration of astronomical data. Large language models, such as Chat-GPT and Falcon, have demonstrated remarkable capabilities in natural language processing and understanding. By leveraging these models’ language generation and comprehension abilities, we aim to develop a simple interface that allows users to interact with astronomy visualization software through natural language commands and queries.
- Natural Language Interaction: Developing an intuitive and user-friendly interface that enables users to interact with astronomy visualization software using natural language commands. This will eliminate the need for users to have prior knowledge of complex software interfaces and programming languages, thus lowering barriers to entry and fostering broader engagement with astronomical data.
- Enhanced Data Exploration: Expanding the capabilities of astronomy archive access software by leveraging large language models’ contextual understanding and information retrieval. This will allow users to ask questions, request specific visualizations, and explore datasets through conversational interactions, enabling more dynamic and exploratory data analysis
Student Attributes
Academic Background
Preferably astronomy/computer science background
Computing Skills
Experience with Unix or linux, python, PyTorch
Training Requirement
Pawsey Training on GPUs
Project Timeline
- Week 1 Inductions and project introduction
- Week 2 Background reading and Initial Presentation
- Week 3 Natural Language Interaction
- Week 4 Natural Language Interaction
- Week 5 Natural Language Interaction
- Week 6 Enhanced Data Exploration
- Week 7 Enhanced Data Exploration
- Week 8 Enhanced Data Exploration
- Week 9 Final Presentation
- Week 10 Final Report
Typical Interface of a Virtual Observatory
