metadata
dataset_info:
features:
- name: paper_id
dtype: uint32
- name: title
dtype: string
- name: paper_url
dtype: string
- name: authors
list: string
- name: type
dtype: string
- name: abstract
dtype: string
- name: keywords
dtype: string
- name: TL;DR
dtype: string
- name: submission_number
dtype: int64
- name: arxiv_id
dtype: string
- name: embedding
list: float64
- name: github
dtype: string
splits:
- name: train
num_bytes: 25554406
num_examples: 3258
download_size: 19637788
dataset_size: 25554406
configs:
- config_name: default
data_files:
- split: train
path: data/train-*