---
dataset_info:
  features:
  - name: source_paper
    dtype: string
  - name: source_section
    dtype: string
  - name: premise
    dtype: string
  - name: target_doi
    dtype: string
  - name: target_paper
    dtype: string
  splits:
  - name: train
    num_bytes: 26085341
    num_examples: 77103
  download_size: 8085835
  dataset_size: 26085341
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---