PARSEME or PARSEME-FR?

We are frequently asked this question: What is the difference between PARSEME and PARSEME-FR? This page tries to a give a concise and precise answer.

PARSEME is a European network on PARsing and Multiword Expressions. Some facts:

• It was funded by the IC1207 COST action in 2013-2017. After the funding period, the network still remains active.
• It is structured into 4 Working Groups, gathering over 200 researchers from 31 countries, working on 30 languages and 6 dialects from 10 language families
• Its main outcomes include:
  • Unified annotation guidelines for verbal MWEs
  • PARSEME multilingual corpus of verbal MWEs, in versions 1.0 (18 languages), 1.1 (20 languages), and 1.2 (14 languages)
  • PARSEME shared task on automatic identification of verbal MWEs, editions 1.0, 1.1, and 1.2.
  • Phraseology and Multiword Expressions (PMWE) - a book series at Language Science Press, Berlin.
• Since the end of the funding period in 2017, the PARSEME follow-up activities have been coordinated by the SIGLEX MWE section.
• The corpus development activities have been ongoing for several years, with regular releases related to the shared tasks, and no dedicated funding.

PARSEME-FR is a French national project, a spin-off of the PARSEME COST action. Basic facts:

• It is funded by the French national ANR research agency in 2016-2020.
• Its consortium gathers 5 French universities
• Its main outcomes include the French Sequoia corpus annotated for all types of MWEs (i.e. not only verbal). The annotation of verbal VMWEs in Sequoia was performed within the PARSEME corpus annotation project and is compatible with its guidelines (see above).

The links between PARSEME-FR and PARSEME are strong. Notably, as far as outcomes are concerned:

• The PARSEME-FR annotation guidelines served as seminal work for the multilingual PARSEME guidelines
• The PARSEME guidelines were used for the annotation of verbal MWEs is Sequoia (but other types of MWEs are annotated in Sequoia, too)
• Consequently, Sequoia (that is, the original corpus and its verbal MWE annotations) is also part of the French subcorpus of the PARSEME corpus

The PARSEME-FR core group members play major roles in the PARSEME network:

• Marie Candito (PARSEME-FR local coordinator at LLF in Paris) is:
  • a coordinator of the PARSEME corpus annotation,
  • a coordinator of the PARSEME shared task (edition 1.2),
  • a language leader for French,
  • the Romance language group leader.
• Eric de la Clergerie (INRIA member in Paris) was:
  • an invited speaker at the PARSEME general meeting in Dubrovnik in 2016.
• Mathieu Constant (PARSEME-FR coordinator) is:
  • the vice-leader of the PARSEME Working Group 3.
• Silvio Ricardo Cordeiro (PARSEME-FR member at LIS, then at LLF) is:
  • a coordinator of the PARSEME shared task (editions 1.0 and 1.1),
  • a language leader for Portuguese.
• Yannick Parmentier (LIFO local coordinator in Orléans, then in Nancy) is:
  • the leader of the PARSEME Working Group 2,
  • an editor of the PMWE book series (until 2020),
  • a co-editor of a PMWE post-PARSEME volume 3 .
• Caroline Pasquer (PARSEME-FR member at LIFAT, then at LIS) is:
  • Author of the VarIDE system, which participated in the PARSEME shared task 2018
• Carlos Ramisch (PARSEME-FR local coordinator at LIS in Marseille) is:
  • a coordinator of the PARSEME corpus annotation,
  • a coordinator of the PARSEME shared task (editions 1.0, 1.1 and 1.2),
  • a language leader for Portuguese,
  • the elected representative of the SIGLEX MWE section (2020-2022),
  • a co-chair of the 2017, 2018 and 2020 editions of the MWE workshop,
  • an editor of the PMWE book series (from 2020),
  • a co-editor of a PMWE post-PARSEME volume 2,
  • a member of the SIGLEX MWE section standing committee (2016-2018).
• Agata Savary (PARSEME-FR local coordinator at LIFAT in Blois/Tours) is:
  • the chair of the PARSEME Management Committee and Steering Committee,
  • the elected representative of the SIGLEX MWE section (2016-2020),
  • a coordinator of the PARSEME corpus annotation,
  • a coordinator of the PARSEME shared task (editions 1.0, 1.1 and 1.2),
  • the Polish language leader,
  • a co-chair of the 2017, 2018, 2019 and 2020 editions of the MWE workshop,
  • an editor of the PMWE book Series,
  • a co-editor of a PMWE post-PARSEME volume 2.
• Nicolas Zampieri and Manon Scholivet (PARSEME-FR interns at LIS) are:
  • Authors of the Veyn system, which participated in the PARSEME shared task 2018