metadata
dataset_info:
features:
- name: bibtex_url
dtype: string
- name: bibtext
dtype: string
- name: abstract
dtype: string
- name: authors
sequence: string
- name: title
dtype: string
- name: id
dtype: string
- name: type
dtype: string
- name: arxiv_id
dtype: string
splits:
- name: train
num_bytes: 8588154
num_examples: 2624
download_size: 4125695
dataset_size: 8588154
configs:
- config_name: default
data_files:
- split: train
path: data/train-*