metadata
dataset_info:
features:
- name: pdf
dtype: string
- name: aff
dtype: string
- name: year
dtype: string
- name: technical_novelty_avg
dtype: float64
- name: video
dtype: string
- name: doi
dtype: string
- name: presentation_avg
dtype: float64
- name: proceeding
dtype: string
- name: presentation
dtype: string
- name: sess
dtype: string
- name: technical_novelty
dtype: string
- name: arxiv
dtype: string
- name: author
dtype: string
- name: site
dtype: string
- name: keywords
dtype: string
- name: oa
dtype: string
- name: empirical_novelty_avg
dtype: float64
- name: poster
dtype: string
- name: openreview
dtype: string
- name: conference
dtype: string
- name: corr_rating_confidence
dtype: float64
- name: corr_rating_correctness
dtype: float64
- name: project
dtype: string
- name: track
dtype: string
- name: rating_avg
dtype: float64
- name: rating
dtype: string
- name: correctness
dtype: string
- name: slides
dtype: string
- name: title
dtype: string
- name: github
dtype: string
- name: authors
dtype: string
- name: correctness_avg
dtype: float64
- name: confidence_avg
dtype: float64
- name: status
dtype: string
- name: confidence
dtype: string
- name: empirical_novelty
dtype: string
splits:
- name: train
num_bytes: 50770708
num_examples: 88224
download_size: 16150671
dataset_size: 50770708
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
size_categories:
- 10K<n<100K
license: apache-2.0
paperlists Dataset
Dataset Details
This dataset contains the list of scientific papers accepted at one of these conferences:
- CVPR (Computer Vision and Pattern Recognition)
- ICCV (International Conference on Computer Vision)
- ECCV (European Conference on Computer Vision)
- CoRL (Conference on Robot Learning)
- ICLR (International Conference on Learning Representations)
- ICML (International Conference on Machine Learning)
- NeurIPS (Conference on Neural Information Processing Systems)
- SIGGRAPH (Special Interest Group on Graphics and Interactive Techniques)
- EMNLP (Conference on Empirical Methods in Natural Language Processing)
From 2006 to the present (depending on the conferences)
Dataset Sources
The data comes from the paperlists repository
Uses
This dataset can be used to produce statistics on papers accepted at TOP-tier AI conferences
Data Collection and Processing
Please note, this is a raw dataset that has undergone little transformation. All fields are original except youtube which was merged into video and id, psid and ssid* which were deleted
Bias, Risks, and Limitations
The data is limited both by conference type and by year They are raw and have a lot of unspecified value