Harsit/scientific-paper-summarizer

Scientific Paper Summarizer

This model is fine-tuned version of google/pegasus-pubmed for scientific paper summarization.

Model description

The model is based on PEGASUS and has been fine-tuned on scientific papers from the ArXiv dataset.

Intended uses & limitations

This model is intended for summarizing scientific papers and research articles.

Training and evaluation data

The model was fine-tuned on a subset of the ArXiv scientific papers dataset.

Training procedure

The model was fine-tuned using the following parameters:

• —Maximum input length: 512
• —Maximum target length: 128
• —Batch size: 4
• —Learning rate: 2e-5
• —Number of epochs: 5

Evaluation results

The model achieved the following ROUGE scores:

• —ROUGE-1: 0.3766
• —ROUGE-2: 0.1260
• —ROUGE-L: 0.2191