Hi Kilian
Hope all has been well.
I'm surprised that people are still "wording around" nowadays. Some suggestions:
1. Can't we rename "MWEs" to "fixed/idiomatic expressions" instead? One can reformulate these as sequences/strings/expressions of various lengths/vocabs in characters. 2. Also, one can interpret these without information/association with any syntactic categories, nouns or verbs etc.. 3. They do just represent lexical info (some reflecting/encoding historico-social habits, though one also should be aware of the ethical aspects of reinforcing some "traditional values"). Perhaps a more sophisticated view of language could help wean practitioners from a mindframe that relies of "linguistic structure(s)" as we've had it thus far (i.e. based on "words" and "sentences")? 4. Re " their meaning often does not result from the direct combination of the meanings of their parts": non-compositionality may be a better description of a more realistic view of language, it should prob be our default expectation (instead of the cherry-picked compositional counterparts).
I think efforts towards mitigating a mental dependency on "words" would be a good direction to pursue, what do you think? Can we get SIGLEX to update in this regard?
Best Ada
On Wed, Feb 8, 2023 at 11:12 AM Kilian Evang via Corpora < corpora@list.elra.info> wrote:
[Apologies for cross-postings]
Call for Papers: Deadline extended
19th Workshop on Multiword Expressions (MWE 2023)
Organized and sponsored by SIGLEX, the Special Interest Group on the Lexicon of the ACL
Full-day workshop collocated with EACL 2023, Dubrovnik, Croatia, May 5 or 6, 2023
Hybrid (on-site & on-line)
NEW: Submission deadline: February 20, 2023
NEW: Invited speakers announced (see below)
NEW: Best paper award (see below)
MWE 2023 website: https://multiword.org/mwe2023/
Multiword expressions (MWEs) are word combinations that exhibit lexical, syntactic, semantic, pragmatic, and/or statistical idiosyncrasies (Baldwin & Kim 2010), such as by and large, hot dog, pay a visit and pull one's leg. The notion encompasses closely related phenomena: idioms, compounds, light-verb constructions, phrasal verbs, rhetorical figures, collocations, institutionalised phrases, etc. Their behaviour is often unpredictable; for example, their meaning often does not result from the direct combination of the meanings of their parts. Given their irregular nature, MWEs often pose complex problems in linguistic modelling (e.g. annotation), NLP tasks (e.g. parsing), and end-user applications (e.g. natural language understanding and MT), hence still representing an open issue for computational linguistics (Constant et al. 2017).
For almost two decades, modelling and processing MWEs for NLP has been the topic of the MWE workshop organised by the MWE section of SIGLEX in conjunction with major NLP conferences since 2003. Impressive progress has been made in the field, but our understanding of MWEs still requires much research considering their need and usefulness in NLP applications. This is also relevant to domain-specific NLP pipelines that need to tackle terminologies most often realised as MWEs. Following previous years, for this 19th edition of the workshop, we identified the following topics on which contributions are particularly encouraged:
MWE processing and identification in specialized languages and domains: Multiword terminology extraction from domain-specific corpora (Bonin et al. 2010) is of particular importance to various applications, such as MT (Semmar & Laib, 2017), or for the identification and monitoring of neologisms and technical jargon (Chatzitheodorou et al, 2021). We expect approaches that deal with the processing of MWEs as well as the processing of terminology in specialised domains can benefit from each other.
MWE processing to enhance end-user applications: MWEs have gained particular attention in end-user applications, including MT (Zaninello & Birch 2020; Han et al. 2021, 2022), simplification (Kochmar et al. 2020), language learning and assessment (Paquot et al. 2019; Christiansen & Arnon 2017), social media mining (Maisto et al. 2017), and abusive language detection (Zampieri et al. 2020; Caselli et al. 2020). We believe that it is crucial to extend and deepen these first attempts to integrate and evaluate MWE technology in these and further end-user applications.
MWE identification and interpretation in pre-trained language models: Most current MWE processing is limited to their identification and detection using pre-trained language models, but we still lack understanding about how MWEs are represented and dealt with therein (Nedumpozhimana & Kelleher 2021; Garcia et al. 2021, Fakharian & Cook 2021), how to better model the compositionality of MWEs from semantics (Moreau et al. 2018). Now that NLP has shifted towards end-to-end neural models like BERT, capable of solving complex tasks with little or no intermediary linguistic symbols, questions arise about the extent to which MWEs should be implicitly or explicitly modelled (Shwartz & Dagan, 2019).
MWE processing in low-resource languages: The PARSEME shared tasks (Ramisch et al. 2020; 2018; Savary et al. 2017), among others, have fostered significant progress in MWE identification, providing datasets that include low-resource languages, evaluation measures, and tools that now allow fully integrating MWE identification into end-user applications. A few efforts have recently explored methods for the automatic interpretation of MWEs (Bhatia, et al. 2018; 2017), and their processing in low-resource languages (Liu & Wang 2020; Kumar et al. 2017). Resource creation and sharing should be pursued in parallel with the development of methods able to capitalize on small datasets (Han et al. 2020).
Through this workshop, we would like to bring together and encourage researchers in various NLP subfields to submit MWE-related research, so that approaches that deal with processing of MWEs including processing for low-resource languages and for various applications can benefit from each other. We also intend to consolidate the converging effects of previous joint workshops LAW-MWE-CxG 2018, MWE-WN 2019 and MWE-LEX 2020, the joint MWE-WOAH panel in 2021, and the MWE-SIGUL 2022 joint session, extending our scope to MWEs in e-lexicons and WordNets, MWE annotation, as well as grammatical constructions. Correspondingly, we call for papers on research related (but not limited) to MWEs and constructions in:
Computationally-applicable theoretical work in psycholinguistics and corpus linguistics;
Annotation (expert, crowdsourcing, automatic) and representation in resources such as corpora, treebanks, e-lexicons, and WordNets (also for low-resource languages);
Processing in syntactic and semantic frameworks (e.g. CCG, CxG, HPSG, LFG, TAG, UD, etc.);
Discovery and identification methods, including for specialized languages and domains such as clinical or biomedical NLP;
Interpretation of MWEs and understanding of text containing them;
Language acquisition, language learning, and non-standard language (e.g. tweets, speech);
Evaluation of annotation and processing techniques;
Retrospective comparative analyses from the PARSEME shared tasks;
Processing for end-user applications (e.g. MT, NLU, summarisation, language learning, etc.);
Implicit and explicit representation in pre-trained language models and end-user applications;
Evaluation and probing of pre-trained language models;
Resources and tools (e.g. lexicons, identifiers) and their integration into end-user applications;
Multiword terminology extraction;
Adaptation and transfer of annotations and related resources to new languages and domains including low-resource ones.
Shared Task
We do not have a shared task this year, but a new release of the PARSEME corpus of verbal MWEs is currently underway. We encourage submission of research papers that include analyses of the new edition of the PARSEME data and improvements over the results for PARSEME 2020 shared task as well as SemEval 2022 task 2 on idiomaticity prediction.
*** Special Track on MWEs in Clinical NLP ***
Pursuing the MWE Section’s tradition of synergies with other communities, this year, we are organizing a joint session with the Clinical NLP workshop for shared papers/poster presentations. Since clinical texts contain an important amount of multiword expressions (e.g. medical terms or domain-specific collocations), a joint session is deemed beneficial for both communities. The goal is to foster future synergies that could address scientific challenges in the creation of resources, models and applications to deal with multiword expressions and related phenomena in the specialised domain of ClinicalNLP. Submissions describing research on MWEs in the specialized domain of ClinicalNLP, especially introducing new datasets or new tools and resources, are welcome. Papers accepted in this track will have the option to present their work in the Clinical NLP workshop at ACL 2023 as well, after being presented at MWE 2023.
Invited Speakers
We are looking forward to invited talks by two amazing speakers:
Leo Wanner, Universitat Pompeu Fabra
TBD
Best paper award
All full papers in the workshop will be considered by the program committee for a best paper award. The decision will be announced in the closing session.
Submission formats
The workshop invites two types of submissions:
archival submissions that present substantially original research in both long paper format (8 pages + references) and short paper format (4 pages + references).
non-archival submissions of abstracts describing relevant research presented/published elsewhere which will not be included in the MWE proceedings.
Paper submission and templates
Papers should be submitted via the workshop's START submission page (https://softconf.com/eacl2023/mwe2023/). Please choose the appropriate submission format (archival/non-archival). Archival papers with existing reviews will also be accepted through the ACL Rolling Review. Submissions must follow the ACL 2023 stylesheet.
Archival papers with existing reviews from ACL Rolling Review will also be considered. A paper may not be simultaneously under review through ARR and MWE. A paper that has or will receive reviews through ARR may not be submitted for review to MWE.
Important Dates
Paper submission: February 20, 2023
ARR paper commitment: March 6, 2023
Notification of acceptance: March 13, 2023
Camera-ready papers due: March 27, 2023
Workshop: May 5 or 6, 2023
All deadlines are at 23:59 UTC-12 (Anywhere on Earth).
Organizing Committee
Program chairs: Marcos Garcia, Voula Giouli, Lifeng Han, Shiva Taslimipoor
Publication chair: Archna Bhatia
Publicity chair: Kilian Evang
Anti-harassment policy
The workshop follows the ACL anti-harassment policy.
Contact
For any inquiries regarding the workshop, please send an email to the Organizing Committee at mweworkshop2023@googlegroups.com. _______________________________________________ Corpora mailing list -- corpora@list.elra.info https://list.elra.info/mailman3/postorius/lists/corpora.list.elra.info/ To unsubscribe send an email to corpora-leave@list.elra.info