La Rochelle Training School - group projects

This page describes the group project organization for the 2nd PARSEME Training School in La Rochelle, 27 June - 1 July, 2016.

Modalities:

  • the invited trainees have submitted manuscripts on their MWE-related scientific activity and challenging issues for discussion (a longer description of some issues is available in dedicated documents)
  • the trainees are divided into 6 groups of 5-8 members each according to their domains of interest
  • each group is supervised by 1-2 senior researchers, her/his role is not to lead the group but to advise it and give feedback, the trainees themselves are in charge of proposals and outcomes
  • each group is assigned 3 challenging issues selected among those sumitted by the trainees
  • 2 sessions of (90 minutes each) are dedicated to the project work; one session on Friday features the presentation of the outcomes (15 minutes per group) - see the program
  • at the beginning each group selects 1-2 reporting persons, who will be in charge of minuting
  • at the end, each group selects 1-2 presenters for the final session
  • each author of a selected discussion issue presents the issue to the group
  • the group work consists in brainstorming, propositing solutions to the issue and describing them
  • the expected outcome consists in several slides (presented in no more than 15 min.) describing at least:
    • the problem statement
    • multilingual examples illustrating the issue, with glosses and translations
    • the solutions proposed by the group
    • conclusions
  • the slides are also sent to parseme_at_nlp_dot_ipipan_dot_waw_dot_pl for on-line publication

Assignments:

  • each trainee reads the manuscripts of all other trainees from his/her group in advance
  • each trainee comes with a A0 poster presenting the contents of his/her manuscript and sends (until 20 June to parseme_at_nlp_dot_ipipan_dot_waw_dot_pl) a poster booster slide for a 2-minute plenary presentation (on Monday)

Groups:

  • [ANN] Annotation issues
    • Supervisor: Victoria Ros√©n
    • Trainees: K√ºbra Adali, Hiwa Asadpour, Natalia Klyueva, Ivana Matas Ivankoviƒá, Carlos Ramisch, Veronika Vincze
    • Discussion issues:
      • How to match a lexicon against a corpus, especially in an agglutinative language? (Variant of the same issue: How to automatically identify inflected forms of MWEs in a corpus, while only their base forms are present in the lexicon?) - submitted by Adali
      • Is it possible to represent semantically idiosyncratic MWEs in (syntactic) treebanks? How? - submitted by Ramisch
      • Status of metaphors wrt. collocations and MWEs - submitted by Vincze
  • [DISC] Discovery issues
    • Supervisor: Fabienne Cap
    • Trainees: Eleri Aedmaa, Chlo√© Artaud, Goranka Blagus Bartolec, Guillaume Chiron, Vittorio Ganfi, Daniela Majchr√°kov√°, Justina Mandravickaite, Shiva Taslimipoor
    • Discussion issues:
      • How to distinguish literal from compositional readings? (Variants of the same issue: How to detect MWEs in context? Differentiating multiword terms from collocations. Diachronic studies for identifying fully from partially grammaticalized items) - submitted by Aedmaa, Blagus, Majchr√°kov√°, Taslimipoor, Ara√∫z, and Ganfi
      • Conflating variants of the same MWE - submitted by Artaud
      • How to apply MWEs for post-OCR correction - submitted by Chiron
  • [LEX] Lexion issues
    • Supervisors: Adam Przepi√≥rkowski & Agnieszka Patejuk
    • Trainees: Eduard Bejƒçek, Made Windu Kesiman, Nikoleta Olexov√°, Lauma Pretkalni≈Üa, Dage S√§rg, Maria Todorova
    • Discussion issues:
      • Encoding the variability of MWEs (Variants of the same issue: How to describe such MWEs that permit some variation to a certain degree in a lexicon? How to describe syntagmatic peculiarities of verbal MWEs and all quantitative and positional changes of the idiom components? While creating a MWE lexicon, how to understand which words are variable and which are fixed in a MWE?) - submitted by Bejƒçek, Todorova, and Pretkalnina
      • Productivity of different elements in a MWE - submitted by S√§rg
      • Lexicon model for MWEs to support transcription of ancient manuscripts - submitted by Kesiman
  • [PARS] Parsing issues
    • Supervisor: Matthieu Constant
    • Trainees: Miryam de Lhoneux, Philippe Grang√©, Verginica Mititelu, Violeta Seretan, Jakub Waszczuk
    • Discussion issues:
      • Impact of transformations of MWE representations (e.g. collapsing into words-with-spaces) in treebanks on parsing accuracy  - submitted by de Lhoneux
      • Which types of MWEs should be handled at the stage of the syntactic parsing?  - submitted by Waszczuk
      • Distinguishing between an idiomatic and a compositional meaning of the same string of words in MT and parsing - submitted by Mititelu
  • [SEM] Semantic issues
    • Supervisor: Koenraad de Smedt
    • Trainees: Carlos Herrero-Zorita, Pilar Le√≥n Ara√∫z, Elpiniki Margariti, Valentina Piunno, Christophe Rigaud
    • Discussion issues:
      • How to correlate free arguments of a MWE with the free arguments of its meaning/paraphrase? - submitted by Margariti
      • Emotional charge of a MWE - submitted by Pirego
      • The role of negation in MWEs (Variants of the same issue: Semantic polarity of syntactically negated MWEs and its links with lexicalization degree. Periphrastic MWEs and their negation) - submitted by Piunno and Herrero
  • [TRANS] Translation issues
    • Supevisors: J√∂rg Tiedemann, Agata Savary
    • Trainees: Mika H√§m√§l√§inen, Jelena Mitrovic, Aleksandar Petrovski, Matƒ´ss Rikters, Beatriz S√°nchez C√°rdenas, Katalin Simk√≥, Georgios Zakis
    • Challenging issues:
      • Pairs of idioms with the same compositional meaning but different idiomatic meanings - submitted by Simko
      • Decoding the semantics of free arguments of MWEs - submitted by Zakis
      • Applying resources of rhetorical figures in machine translation - submitted by Mitrovic