metadata
dataset_info:
features:
- name: id
dtype: string
- name: proceedings
dtype: string
- name: type
dtype: string
- name: arxiv_id
dtype: string
- name: title
dtype: string
- name: authors
sequence: string
- name: paper_page
dtype: string
- name: conference_name
dtype: string
splits:
- name: train
num_bytes: 7882807
num_examples: 26607
download_size: 3569180
dataset_size: 7882807
configs:
- config_name: default
data_files:
- split: train
path: data/train-*