In-Context Language Modelling For Paper Source Tracing
Keywords: In-Context, Source Tracing, Fine-tuning
Abstract: With the rapid advancement of technology, the number of academic papers is surging at an alarming rate. Millions of papers are published globally every year, and the number continues to climb. It is becoming increasingly challenging for researchers to trace the pulse of technological development from such a vast ocean of literature. For this reason, we propose a new approach: a comprehensive analysis using the contextual information of a paper as well as other literature it cites. This approach demonstrates significant results on a test set and a rapid inference process, providing researchers with an efficient tool for literature screening and analysis.
Submission Number: 24
Loading