MuP: Multi-Perspective Scientific Document Summarization Dataset from Peer Review Summaries (SDP 2022 Shared Task, Allen Institute for AI)

Provider
Allen Institute for AI and MuP 2022 organizers (Guy Feigenblat, Arman Cohan et al.)
License
ODC-By-1.0 (source)
Stored at
Amazon S3, s3://ai2-s2-research-public/mup-dataset (us-west-2)
Access
Free and public. No account with the data provider is needed.
Category
Natural Language Processing
Source
https://github.com/allenai/mup

About this dataset

MuP is a scientific document summarization corpus with multiple gold summaries per paper, built from peer reviews: reviewers' opening paragraphs that summarize a paper's contributions from their point of view are used as summaries, so each paper has several perspectives. It was released for the Multi Perspective Scientific Document Summarization shared task at the Scholarly Document Processing workshop (SDP at COLING 2022). Files: mup.zip (405 MB, train and validation) with simplified CSV files (training.csv, validation.csv, the first 2,000 tokens of each paper) and JSON-lines files with the full paper text, section structure and metadata; testing_with_paper_release.jsonl.zip and test-release-simple.csv.zip for the test set. Used for multi-reference summarization, long-document summarization of scientific papers and evaluation of summary diversity. Baselines and the leaderboard are in allenai/mup. License: ODC-BY. Cite the MuP shared task overview from SDP 2022 (organizers Guy Feigenblat, Arman Cohan and colleagues).

How to access this dataset in Python

import boto3
from botocore import UNSIGNED
from botocore.client import Config
s3 = boto3.client('s3', region_name='us-west-2', config=Config(signature_version=UNSIGNED))
s3.list_objects_v2(Bucket='ai2-s2-research-public', Prefix='mup-dataset/', Delimiter='/')

Lists the files with anonymous access. Runs on any machine, with no cloud account.

Data structure

Files

Example file: mup-dataset/mup.zip

Analyze this dataset without downloading it

On Scigantic, MuP: Multi-Perspective Scientific Document Summarization Dataset from Peer Review Summaries (SDP 2022 Shared Task, Allen Institute for AI) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Natural Language Processing datasets