Finetuning Large Language Models to perform summarization on research papers.
A base T5 auto-regressive model was finetuned on this dataset of research papers paired with plain language summaries to generate readable summaries from technical abstracts.
The model achieved the following evaluation metrics:
- ROUGE-1: 0.403590
- ROUGE-2: 0.124948
- ROUGE-L: 0.214003