About this Abstract |
| Meeting |
2025 TMS Annual Meeting & Exhibition
|
| Symposium
|
Novel Strategies for Rapid Acquisition and Processing of Large Datasets from Advanced Characterization Techniques
|
| Presentation Title |
A Retrieval-Augmented Generation Application in the Dental Composites Space |
| Author(s) |
Wade Smallwood, Ramsey Issa, Hasan Sayeed, Taylor Sparks |
| On-Site Speaker (Planned) |
Wade Smallwood |
| Abstract Scope |
Retrieval-Augmented Generation (RAG) is a recent development in the NLP space that addresses the critical limitations of factual inaccuracies and output hallucination inherent to off-the-shelf Large Language Models. This is accomplished by integrating a generative language model with a robust retrieval mechanism. In this work, we demonstrate how scientists can systematically extract data in data-friendly formats from published works, facilitating the systematic organization and development of novel databases from existing literature, along with a more context aware interactive language model to aid any future exploration in the respective field. |
| Proceedings Inclusion? |
Planned: |
| Keywords |
Machine Learning, Biomaterials, Composites |