[CfP] First Shared Task on Multi-Perspective Scientific Document Summarization (MuP) at 3rd SDP @ COLING 2022

51 views
Skip to first unread message

Tirthankar Ghosal

unread,
May 30, 2022, 8:49:59 AM5/30/22
to Tirthankar Ghosal

***Call for Participation***


***First Shared Task on Multi-Perspective Scientific Document Summarization (MuP)***


Website: https://github.com/allenai/mup 


Generating summaries of scientific documents is known to be a challenging task. The majority of existing work in summarization assumes only one single best gold summary for each given document. Having only one gold summary negatively impacts our ability to evaluate the quality of summarization systems, as writing summaries is a subjective activity. At the same time, annotating multiple gold summaries for scientific documents can be extremely expensive as it requires domain experts to read and understand long scientific documents. This shared task will enable exploring methods for generating multi-perspective summaries. We introduce a novel summarization corpus, leveraging data from scientific peer reviews to capture diverse perspectives from the reader's point of view (each paper has multiple summaries reflecting multiple perspectives of the reader).


The MuP shared task is a part of the 3rd Scholarly Document Processing (SDP) workshop at COLING 2022. https://sdproc.org/2022/ 


More details on the shared task and the corresponding dataset can be found on: https://github.com/allenai/mup 


***Please fill in this form to participate in the shared task*** 

https://forms.gle/K2UECKvmghzDHUpo7 


The leaderboard for the shared task will be announced soon on the website.


Shared Task Timelines


Training Data Release: May 10, 2022 

Test Data Release: June 30, 2022 

Evaluation Period: July 1 - July 15, 2022 

System Description Papers Due: August 1, 2022

Reviews Notification: August 15, 2022 

Camera-Ready Papers Due: September 5, 2022

Event at SDP @ COLING 2022: October 16/17, 2022


MuP 2022 Organizers


  1. Guy Feigenblat - Piiano, Israel

  2. Arman Cohan - AI2, US

  3. Tirthankar Ghosal - ÚFAL, Charles University, Czechia

  4. Michal Shmueli-Scheuer - IBM Research AI, Israel


--

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Tirthankar Ghosal

Researcher at UFAL, Charles University, CZ

https://member.acm.org/~tghosal

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Tirthankar Ghosal

unread,
Jun 26, 2022, 2:36:38 AM6/26/22
to Tirthankar Ghosal
Reply all
Reply to author
Forward
0 new messages