- Corpora - ELRA lists

CD-MAKE 2023 special session on “Multi-Perspectivist Data and Learning”
by Valerio Basile 08 Mar '23

08 Mar '23

Dear colleagues, We are happy to announce the call for papers for our one-day special session on “Multi-Perspectivist Data and Learning 2023” at the upcoming CD MAKE 2023 conference: https://cd-make.net/special-sessions/multi-perspectivist-data-and-learning/ *************************************************** Description, scope and aims Many Artificial Intelligence applications are based on supervised machine learning (ML), which ultimately grounds on manually annotated data. The annotation process (i.e., ground-truthing) is often performed in terms of a majority vote and this has been proved to be often problematic, as highlighted by recent studies on the evaluation of ML models. Recently, a different paradigm for ground-truthing has started to emerge, called data perspectivism, which moves away from traditional majority aggregated datasets, towards the adoption of methods that integrate different opinions and perspectives within the knowledge representation, training, and evaluation steps of ML processes, by adopting a non-aggregation policy. This alternative paradigm obviously implies a radical change in how we develop and evaluate ML systems: such ML systems have to take into account multiple, uncertain, and potentially mutually conflicting views. This obviously brings both opportunities and difficulties: novel models or training techniques may need to be designed, and the validation phase may become more complex. Nonetheless, initial works have shown that data perspectivism can lead to better performances, and could also have important implications in terms of human-in-the-loop and interpretable AI, as well as in regard to the ethical issues or concerns related to the use of AI systems. The scope of this special session is to attract contributions related to the management of subjective, crowd-sourced, multi-perspective, or otherwise non-aggregated data in ground-truthing, machine learning, and more generally artificial intelligence systems. Invited contributions: full research papers and research in progress papers. *************************************************** Topics of interest: - Subjective, uncertain, or conflicting information in annotation and crowdsourcing processes; - Limits and problems with standard data annotation and aggregation processes; - Theoretical studies on the problem of learning from multi-rater and non-aggregated data; - Participation mechanisms/incentives/gamification for rater engagement and crowdsourcing; - Ethical and legal concerns related to annotation and aggregation processes in ground-truthing; - Creation and documentation of multi-rater and non-aggregated datasets and benchmarks; - Development of ML algorithms for multi-rater and non-aggregated data; - Techniques for the evaluation of ML systems based on multi-rater and non-aggregated data; - Applications of data perspectivism and non-aggregated data to eXplainable AI, human-in-the-loop AI and algorithmic fairness; - Experimental and application studies of ML/AI systems on multi-rater and non-aggregated data, in possibly different application domains (e.g. NLP, medicine, legal studies, etc.) *************************************************** Important dates: Submission Deadline March 27, 2023 (AoE) Author Notification June 01, 2023 Proceedings Version June 22, 2023 (AoE) Conference August 29 – September 01, 2023 *************************************************** Special Session Chairs: Federico Cabitza (University of Milano-Bicocca, Italy) Andrea Campagner (University of Milano-Bicocca, Italy) Valerio Basile (University of Turin, Italy) Program Committee (provisional): Nahuel Costa Cortez, University of Oviedo Elisa Leonardelli, Fondazione Bruno Kessler (FBK) Julian Lienen, Paderborn University Gavin Abercrombie, Heriot-Watt University Simona Frenda, University of Turin Marília Barandas, Fraunhofer Portugal AICOS Duarte Folgado, Fraunhofer Portugal AICOS Barbara Plank, Ludwig Maximilian University of Munich Tommaso Caselli, Rijksuniversiteit Groningen *************************************************** Related readings [1] Cabitza, F., Campagner, A., Basile, V. (2023) Toward a Perspectivist Turn in Ground Truthing for Predictive Computing Proceedings of the AAAI Conference on Artificial Intelligence (extended preprint at: https://arxiv.org/pdf/2109.04270.pdf) [2] V. Basile (2020) It’s the End of the Gold Standard as we Know it. On the Impact of Pre-aggregation on the Evaluation of Highly Subjective Tasks Proceedings of the AIxIA 2020 Discussion Papers Workshop [3] F. Cabitza, A. Campagner, L. M. Sconfienza (2020) As if sand were stone. New concepts and metrics to probe the ground on which to build trustable AI BMC Medical Informatics and Decision Making [4] Plank, B. (2022). The 'Problem' of Human Label Variation: On Ground Truth in Data, Modeling and Evaluation. arXiv preprint arXiv:2211.02570.

1 1

Ph.D. Studentship Essex University, UK
by Ravi Shekhar 07 Mar '23

07 Mar '23

Dear All, I am soliciting expressions of interest for a fully funded Ph.D. Studentship at the School of Computer Science and Electronic Engineering, Essex University, UK, starting in October 2023. The scholarship will cover three years of home/international tuition fees and a living allowance. You should have a master's degree with a strong background in Natural Language Processing, Machine Learning, and/or Linguistics. You will have the opportunity to define your research agenda in the following broad research themes: - Multi-modal Conversational Agent - Low-resource NLP - Combining social media analysis with NLP Please send me (r.shekhar(a)essex.ac.uk) your updated CV and a brief explanation of your research interest with the subject line “Ph.D. Studentship(2023): Your Name.” -- Ravi Shekhar

1 0

Call for Papers: IWCS workshop on Semantics in Organisational Communication
by Matthew Purver 07 Mar '23

07 Mar '23

CFP: SemInOrgCom: Semantics in Organisational Communication https://sodestream.github.io/seminorgcom/ June 20th, Nancy, France Co-located with IWCS 2023 Important dates: Regular and non-archival submissions: * April 14th Submission deadline * May 12th Notification of acceptance * June 9th Camera ready deadline for regular papers Hackathon: * March 8th – May 31st Participants working on Hackathon * March 31st First meeting with Hackathon participants * May 2nd Second meeting with Hackathon participants * June 7th Submission deadline for system description papers * June 9th Camera ready deadline for hackathon system description papers Workshop: * 20th Jun Workshop date Keynote speakers: Dr. Colin Perkins (University of Glasgow; chair, Internet Research Task Force) Prof. Magda Osman (University of Cambridge) Workshop description: Interaction and communication are at the heart of every organisation, from small to large, and take many forms: email, group messaging applications, face-to-face and online meetings of various sizes, and others. Insights into how people express complex issues, discuss their own and others’ intentions and make decisions could help make these processes more efficient and/or transparent and lead to a range of assistive tools. However, the group interaction involved is often at a scale between the small scales usually assumed in computational semantics or dialogue modelling, and the very large scales usually studied in social networks. The organisational nature also brings important factors that affect language and the meaning expressed or understood – explicit or implicit hierarchy, shared or disputed goals, and social groupings with competitive or collaborative agendas – well known in other disciplines but not often taken into account in computational semantics. The computational linguistics community has looked at various relevant phenomena and tasks (e.g., meeting summarization, intention detection, intention detection, argument mining, agreement/disagreement detection, persuasiveness detection), and some relevant datasets have been produced (e.g., the Enron email dataset). However, there are still relatively few attempts and few resources or approaches to semantics in organisational communication in general. This workshop aims to fill this gap to model, analyse and understand overall organisational communication, and encourage collaboration between researchers from diverse backgrounds, including computational linguistics, organisational psychology, and computational social science. Main topics: We welcome work broadly in the area of natural language processing, computational linguistics, computational social science, sociolinguistics, organisational psychology, and related fields with the aim of better understanding organisational communication. Cross-disciplinary collaborations between computer scientists and other social scientists in order to reach richer insights are especially welcome. We also encourage contributions that address multilingual settings as well as low-resource languages. The workshop topics include but are not limited to: * Summarization of meetings and other organisational communication * Models of argument, (dis)agreement and decision-making * Analysis of influence, persuasiveness and power relations * Effects of organisational culture and hierarchy * Communication across different modalities and timescales * Differences between organisational communication and other forms of communication * Datasets and annotation schemas for organisational communication * Social network analysis in organisations as applied to communication * Diachronic analysis of organisational communication * Application and adaptation of NLP models to organisational communication. Format: Regular submissions (long and short) Authors are invited to submit full papers of up to 8 pages of content and short papers of up to 4 pages of content, with unlimited pages for references. Accepted papers will be given an additional page of content to address reviewer comments and will be published in the ACL Anthology. Previously published papers cannot be accepted. Dual submissions are allowed; papers that are currently undergoing review at other venues are welcome but must declare this on submission. Non-archival submissions We welcome two types of non-archival submissions. First, you can submit an extended abstract of work not published elsewhere. These can include position papers, or early-stage work that would benefit from peer feedback. Second, work previously accepted/published elsewhere, along with details about the venue or journal where it is accepted, and a link to the archived version, if available. In both cases there are no page limits, or style/anonymity requirements, and the submissions will be reviewed only for the fit to the workshop theme. Papers accepted as non-archival will be given an opportunity to present the work at the workshop but will not be published in the ACL Anthology (they will be available on the workshop website). Hackathon submissions An active, experimentation-based track where hackathon-type online activities precede the workshop, and teams/individuals present their work in the workshop. The hackathon organisers will provide data, task suggestions, and periodic feedback. Though, participants are free to work on any relevant task or dataset during their hackathon project. Hackathon activities are by design online, while the rest of the workshop will be in person. Hackathon participants are invited (but not required) to submit a system description paper (up to 4 pages + unlimited pages for references); authors will be able to choose whether these are published in the ACL anthology. Journal special issue After the workshop, we will explore the possibility of inviting selected authors to submit a paper to a special issue of the Dialogue & Discourse journal. The journal submissions would undergo further review, and the paper should be substantially different from the original work. Submission instructions: Similar to IWCS, regular submissions should be fully anonymous to ensure double-blind reviewing. All submissions should follow the IWCS conference template (see https://iwcs2023.loria.fr/call-for-papers/) Submission link: https://softconf.com/iwcs2023/seminorgcom/ Program committee: Ignacio Castro, Queen Mary University of London Goran Glavaš, University of Würzburg Patrick Healey, Queen Mary University of London Mladen Karan, Queen Mary University of London Stephen McQuistin, University of Glasgow Paul Piwek, Open University Matthew Purver, Queen Mary University of London and Jožef Stefan Institute Ravi Shekhar, University of Essex Gareth Tyson, Hong Kong University of Science of Science and Technology Andreas Vlachos, University of Cambridge Ivan Vulić, University of Cambridge (more TBA) -- Matthew Purver - http://www.eecs.qmul.ac.uk/~mpurver/ Computational Linguistics Lab - http://compling.eecs.qmul.ac.uk/ Cognitive Science Research Group - http://cogsci.eecs.qmul.ac.uk/ School of Electronic Engineering and Computer Science Queen Mary University of London, London E1 4NS, UK *My working days for QMUL are Monday-Wednesday; responses to mail on other days may be delayed.*

1 0

Jobs at University of Leeds, UK
by Eric Atwell 07 Mar '23

07 Mar '23

University of Leeds hopes to recruit 130+ new staff, including: Lecturers in Computer Science (Software Engineering)<https://jobs.leeds.ac.uk/vacancy.aspx?ref=EPSCP1123> deadline 19/03 Lecturers in Computer Science (System Architecture)<https://jobs.leeds.ac.uk/vacancy.aspx?ref=EPSCP1124> 19/03 Lecturers in Computer Science (Theoretical Computer Science)<https://jobs.leeds.ac.uk/vacancy.aspx?ref=EPSCP1125> 19/03 Lecturer in Electronic and Electrical Engineering<https://jobs.leeds.ac.uk/vacancy.aspx?ref=EPSEE1092> 27/03 Software Development Scientist<https://jobs.leeds.ac.uk/vacancy.aspx?ref=ENVEE1637> 27/03 Research Engineer in Internet-of-Things<https://jobs.leeds.ac.uk/vacancy.aspx?ref=EPSEE1093> 5/04 Research Software Engineer<https://jobs.leeds.ac.uk/vacancy.aspx?ref=EPSCP1122> 5/04

1 0

PhD position on SocialNLP at University of Tartu
by Somnath Banerjee 07 Mar '23

07 Mar '23

Dear all, I am looking for motivated students for PhD (4-year) on SocialNLP. SocialNLP is a new interdisciplinary area of natural language processing (NLP) and social computing. The scope of doctoral thesis proposals (but is not restricted to) is based on the following topics: - Content analysis on Social Media (Named Entity Recognition, Sentiment analysis, etc.) - Information Disorder on Social Media (Fake news, Cyberbullying, Hate Speech, misinformation, etc.) - Disaster Management Using Social Media - Explainable AI in SocialNLP Why the University of Tartu<https://ut.ee/en>? - The University of Tartu is one of the oldest universities (founded in 1632) in Northern and Eastern Europe. The University of Tartu is known for its high-quality studies and research. The unique intellectual atmosphere and fine living and working environment that characterise Tartu have attracted prominent scientists and experts for almost 390 years. - As of 2023, the University of Tartu belongs to the top 1% of the world's universities. - The University of Tartu is the strongest research university in Estonia. We belong to various networks of renowned research universities, such as ENLIGHT, the Coimbra Group, LERU, etc. - Here, 13 000 students and 3000 staff members study and work in a diverse range of fields connected by the power of curiosity. - Estonia is famous for its digital society and has the highest number of startups per capita in Europe. Estonia is famous for its digital innovations, and it is home to many startups like Skype, Transferwise, and Bolt to name a few. Scholarship: - Competitive scholarship which covers health insurance (as per the UT rules and regulations) - Additional benefits: Travel grant for attending conferences and workshops, etc. If you are interested, please drop me an email (somnath.banerjee(a)ut.ee) with the followings: - CV which must include: - final year undergraduate/graduate project (brief description: max 150 words) - pursued relevant subjects at undergraduate and/or Master level (if any) - relevant work experience (if any) - publications (if any) - The name and email contacts of academic referees(optional) Please Note: - Master-level students, who are in their final semester, also can apply. - Applications will be considered until the position is filled. However, you are encouraged to apply early as the processing of the applications will start soon after receiving them. - Upon initial selection based on CV, additional documents will be asked for further processing, such as motivation later, etc. For queries: If you need more information, please contact me with the subject line "SocialNLP: PhD position queries". Kind regards, Somnath Banerjee Lecturer (~ Assistant Professor), Chair of Data Science, Institute of Computer Science, University of Tartu, Narva mnt 18 room 3063 51009 Tartu, ESTONIA webpage: http://www.ut.ee/~somnath/

1 0

Final Call for Papers @ Fourth Biennial Conference on Language, Data and Knowledge (LDK 2023)
by Fahad Khan 07 Mar '23

07 Mar '23

Call for Papers @ Fourth Biennial Conference on Language, Data and Knowledge (LDK 2023) Dates: 12–13 September 2023 (Workshops/Tutorials), 14–15 September 2023 (Main Conference) Location: Vienna, Austria Website: http://2023.ldk-conf.org Submission Deadline: 27 March 2023 (revised, was previously 10 March 2023) Submission page: https://openreview.net/group?id=LDK/2023/Conference ============== We invite submissions to the fourth biennial conference on Language, Data and Knowledge (LDK 2023) to be held in Vienna, Austria in September 2023. This conference aims to bring together researchers from across different disciplines concerned with the acquisition, treatment, curation and use of language data in the context of data science and knowledge-based applications. This edition builds upon the success of the inaugural event held in Galway, Ireland in 2017, the second LDK in Leipzig, Germany in 2019, and the third LDK in Zaragoza, Spain in 2021. Invited speakers We are happy to announce Diana Maynard (University of Sheffield), Ruben Verborgh (Ghent University), and Ruth Wodak (Lancaster University/University of Vienna), as keynote speakers for LDK 2023. Paper submission We welcome submissions of relevance to the topics listed below. Submissions can be in the form of: Long papers: 9–12 pages; Short papers: 4–6 pages. All submission lengths are given including references. Accepted submissions will be published by ACL in an open-access conference proceedings volume, free of charge for authors. The ACL templates should therefore be used for all conference submissions. As the reviewing process is single-blind, submissions should not be anonymised. Papers should be submitted via OpenReview at the following address: https://openreview.net/group?id=LDK/2023/Conference All articles must represent original work: when submitted, the submission must not have been previously published*, and the material in it must not have been/be submitted for review at another journal or conference while under review at LDK 2023. *This excludes papers on preprint archives, such as arXiv, which we do not consider to have been previously published. The conference will be hybrid (face-to-face and remote). Note that at least one author of each accepted paper must register to present the paper at the conference (either remotely or on-site). There will be no registration fee administered for participating in LDK 2023. Presentation format Accepted submissions will be selected for oral or poster presentation based on recommendations from the reviewers. This decision will not reflect any difference in the quality of the papers, and there will be no distinction between oral and poster presentations in the published proceedings. Authors of accepted short papers or posters are welcome to present their work as a demo in addition to the regular presentation. Topics Relevant topics for the conference include, but are not limited to, the following fields: Language Data Language data construction and acquisition Language data annotation FAIR data practices for language data Language data portals and metadata about language data Organisational and infrastructural management of language data Multilingual, multimedia and multimodal language data Evaluation, provenance and quality of language data Visualisation of language data Standards and interoperability of language data Legal aspects of publishing language data Under-resourced languages e-Lexicography Semantic processing Knowledge Graphs Linguistic linked data and the multilingual semantic web Ontologies, terminologies, wordnets, framenets and related resources Information and knowledge extraction (taxonomy extraction, ontology learning) Data, information and knowledge integration across languages (cross-lingual) ontology alignment Entity linking and relatedness Linked data profiling Knowledge representation and reasoning Knowledge graphs for corpora processing and analysis Applications for Language, Data and Knowledge Question answering and semantic search Text analytics on big data NLP for language documentation and preservation Speech recognition and synthesis Spoken language processing Semantic content management Computer-aided language learning Natural language interfaces to big data Knowledge-based NLP Deep learning and machine learning for and on LLOD Other applications Use Cases in Language, Data and Knowledge Contributions are welcome where the topics above - and others within the scope of Language, Data and Knowledge - are applied to domain-specific use cases, including but not limited to: social sciences and humanities, legal, life sciences, FinTech, cybersecurity. Organising committee Conference Chairs Jorge Gracia – University of Zaragoza John P. McCrae – University of Galway Program Chairs: Sara Carvalho – University of Aveiro | NOVA CLUNL Anas Fahad Khan – Institute for Computational Linguistics “A. Zampolli” Workshop and Tutorial Chairs: Ana Ostroški Anić – Institute of Croatian Language and Linguistics Blerina Spahiu – University of Milano-Bicocca Local Organisers: Dagmar Gromann – University of Vienna Barbara Heinisch – University of Vienna Proceedings Chair: Ana Salgado – NOVA CLUNL | Lisbon Academy of Sciences Important Dates (Revised) 27 March 2023 New paper submission deadline 5 May 2023 Notification 2 June 2023 Camera-ready submission deadline 12–13 September 2023 Pre-conference events 14–15 September 2023 Main conference All deadlines refer to anywhere-on-earth time. Program Committee http://2023.ldk-conf.org/program-committee/ Workshops and tutorials You can check the list of accepted workshops and tutorials at http://2023.ldk-conf.org/workshops-tutorials/

1 0

Release of the International Corpus of English - The Gibraltarian Component (ICE-GBR)
by Cristina Suárez Gómez 06 Mar '23

06 Mar '23

Dear colleagues, We are pleased to announce that the Gibraltarian written component of the International Corpus of English (ICE-GBR) is now available to all members of the research community. ICE-GBR was compiled by members of the ViEW research group<http://view0.webs.uvigo.es/> and contains 400,000 words of written text in Gibraltarian English. We expect the spoken component to be out soon, since we have recently managed to establish successful collaboration with our colleagues from different Gibraltarian Institutions (Ministry of the Environment, Sustainability, Climate Change, Heritage and Culture of Gibraltar, Gibraltar Cultural Services, University of Gibraltar, Gibraltar College, Gibraltar Chamber of Commerce, etc.). We truly appreciate their selfless help and that of the Gibraltarian writer and project member M.G. Sanchez. The corpus with a description of the contents can now be accessed at view0.webs.uvigo.es/ice-gibraltar<http://view0.webs.uvigo.es/ice-gibraltar>, via links to Zenodo and to the ICE hub in Zurich. The ViEW research group

1 0

SIGDIAL/INLG 2023: Second Call for Workshop & Tutorial Proposals (Submit by March 24)
by Svetlana Stoyanchev 06 Mar '23

06 Mar '23

The 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDial 2023) and the 16th International Natural Language Generation Conference (INLG 2023) will be held jointly in Prague on September 11-15, 2023. We now welcome the submission of workshop and tutorial proposals, which will take place on September 11 and 12 before the main conference. We encourage submissions of proposals on any topic of interest to the discourse, dialogue, and natural language generation communities. This program is intended to offer new perspectives and bring together researchers working on related topics. We especially encourage the sessions that would bring together researchers from SIGDial and INLG communities. Topics of interest include all aspects related to Dialogue, Discourse and Generation including (but not limited to) annotation and resources, evaluation, large language models, adversarial and RL methods, explainable/ethical AI, summarization, interactive/multimodal/situated/incremental systems, data/knowledge/vision-to-text, and applications of dialogue and NLG. The proposed workshops/tutorials may include a poster session, a panel session, an oral presentation session, a hackathon, a generation/dialogue challenge, or a combination of the above. Workshop organizers will be responsible for soliciting, reviewing, and selecting papers or abstracts. The workshop papers will be published in a separate proceedings. Workshops may, at the discretion of the SIGDial/INLG organizers, be held as parallel sessions. Submissions Workshop and Tutorial proposals should be 2-4 pages containing: title, type (workshop or tutorial), a summary of the topic, motivating theoretical interest and/or application context; a list of organizers and sponsors; duration (half-day or full-day), and a requested session format(s): poster/panel/oral/hackathon session. Please include the number of expected attendees. The workshop proposals will be reviewed jointly by the general chair and program co-chairs. Links Those wishing to propose a workshop or tutorial may want to look at some of the sessions organized at recent SIGDial meetings: Natural Language in Human Robot Interaction (NLiHRI 2022) <https://2022.sigdial.org/call-for-papers-nlihri/> NLG4health 2022 <https://nlg4health.uvt.nl/> SummDial 2021 <https://elitr.github.io/automatic-minuting/summdial.html> SafeConvAI 2021 <https://sites.google.com/view/safety4convai/home> RoboDIAL 2022 <https://robodial.github.io/> BigScience Workshop: LLMs 2021 <https://bigscience.huggingface.co/> Interactive Natural Language Technology for Explainable Artificial Intelligence 2019 <https://sites.google.com/view/nl4xai2019/> https://www.inlg2019.com/workshop https://www.sigdial.org/files/workshops/conference18/sessions.htm https://inlg2018.uvt.nl/workshops/ Important Dates Mar 24, 2023: Workshop/Tutorial Proposal Submission Deadline April 14, 2023: Workshops/Tutorials Notifications The proposals should be sent to conference(a)sigdial.org

1 0

CFP TAL journal - special issue "Robustness and limitations of natural language processing models"
by Aurélie Névéol 06 Mar '23

06 Mar '23

[sent on behalf of the special issue editors] Hello all, apologies for the cross-posting. This special issue of the TAL journal calls for papers on the topic of the robustness and limitations of natural language processing (NLP) models (written text). The deadline for submission of articles is 12 May 2023 contrary to what had been announced earlier by mistake. The corrected call is available below. Kind regards, Caio Corro ----- **Robustness and limitations of natural language processing models** This special issue of the TAL journal calls for papers on the topic of the robustness and limitations of natural language processing (NLP) models (written text). Call for submission: http://tal-64-2.sciencesconf.org/ Paper submission deadline : May 12th, 2023 THE TAL JOURNAL Since 1960, ATALA has been publishing the international journal Traitement Automatique des Langues (formerly La Traduction Automatique, then TA Informations), with the support of the CNRS. This journal is published three times a year. The journal is open-access: free to submit, free to publish and free to read. Published articles will be available on the ATALA* website and on the ACL Anthology**. Manuscripts may be submitted in English or French. French-speaking authors are requested to submit their contributions in French. * https://www.atala.org/revuetal ** aclanthology.org/ GUEST EDITORS - Caio Corro (Université Paris-Saclay, CNRS, LISN) - Gaël Lejeune (Sorbonne Université, STIH) - Vlad Niculae (Language Technology Lab, University of Amsterdam) FOCUS OF THE ISSUE Machine learning methods have made it possible to achieve spectacular results on numerous NLP tasks and benchmarks, giving the impression that many problems related to NLP are "solved" or soon to be solved. Nevertheless, it is still an open question as whether or not these models are really effective, or even usable, in non-controlled environments. This thematic issue of the TAL journal aims at questioning the robustness and limits of modern NLP models, in particular related to the following aspects: 1. "Non-standard" data: the use of models on non-standard data, that is data presenting variations with respect to language (diachronic language variation, diatopic variations, variation in word order, code-switching, user-generated content, inconsistent spelling, accidentally noisy data due to pre-processing, incomplete data, presence of specialized domain vocabulary...). 2. Out-of-domain data: the use of models on data from a different domain that the one seen during training; 3. Linguistic structures unseen during training: compositional generalization [1], structural generalization [2], or generalization related to gender bias [3], among others. Relevant topics for this thematic issue include, but are not limited to, the following areas : - identification and evaluation of linguistic phenomena that are problematic for neural networks and other NLP systems; - analysis and correction of error propagation in NLP systems (cascading error analysis); - feedback on the use of NLP systems that have been found to be non-functional on specific types of data; - critical analysis of datasets used for learning or evaluation; - construction of datasets to evaluate robustness with respect to linguistic variations; - data augmentation to improve robustness; - out-of-domain adaptation or learning with domains that are underrepresented in the data; - neural architectures or training methods that improve robustness. All standard NLP tasks can be considered. Works on other languages than French and English are warmly welcomed. IMPORTANT DATES - Paper submission deadline : May 12th, 2023 via https://tal-64-2.sciencesconf.org/ - Notification to the authors after first review : July 2023 - Notification to the authors after second review : November 2023 - Publication : February 2024 LANGUAGE Articles are written in English or French. Submissions in English are accepted only if one of the co-authors is not French speaking. SUBMISSION FORMAT The length of the papers must be between 20 and 25 pages. The TAL journal has a double-blind review process. It is necessary to anonymize the article, the name of the file, and to avoid self-references. Style sheets are available on the journal's website: https://www.atala.org/content/instruction-authors-style-files-0 Authors are invited to submit their paper by clicking on the menu "Paper submission" (PDF article" menu (PDF format). To do so, you will need to have an account on the sciencesconf platform ( http://www.sciencesconf.org ). Click on "create account" next to the "Connect" button at the top of the page. To submit, come back to the page https://tal-64-2.sciencesconf.org/, connect to your account and upload your submission. CONTACT tal-64-2(a)sciencesconf.org REFERENCES [1] COGS: A Compositional Generalization Challenge Based on Semantic Interpretation (Najoung Kim, Tal Linzen), EMNLP 2020 https://aclanthology.org/2020.emnlp-main.731/ [2] Structural generalization is hard for sequence-to-sequence models (Yuekun Yao, Alexander Koller), EMNLP 2022 https://arxiv.org/abs/2210.13050 [3] Evaluating Gender Bias in Machine Translation (Gabriel Stanovsky, Noah A. Smith, Luke Zettlemoyer), ACL 2019 https://aclanthology.org/P19-1164/ -- Aurélie Névéol, PhD, HdR LISN-CNRS Bâtiment 507, bureau A111 Rue du belvédère Université Paris-Saclay 91403 ORSAY France Tel: +33 (0)1 69 15 81 59

1 0

MultiGED-2023 shared task: last call, data updates and extension
by Elena Volodina 06 Mar '23

06 Mar '23

Updates to MultiGED-2023 shared task on multi-lingual error detection (https://github.com/spraakbanken/multiged-2023): (1) Train and dev data have been updated following reports from the participants. Please, re-download to use the latest version! (https://github.com/spraakbanken/multiged-2023). Apologies for the last-minute updates! Discussion on the nature of data corrections is on the google group: https://groups.google.com/g/multiged-2023 Dev data on CodaLab is updated accordingly. (2) The test phase is now open on CodaLab: (https://codalab.lisn.upsaclay.fr/competitions/9784) (3) In connection to (1), we are extending the deadline for system submissions till Monday, 13 March, 23:59:59 UTC Read more about the shared task here: https://github.com/spraakbanken/multiged-2023 Best regards, Computational SLA working group (https://spraakbanken.gu.se/en/compsla) ___________________ Elena Volodina, PhD, Docent https://spraakbanken.gu.se/en/about/staff/elena Life is like a mirror. Smile at it and it smiles back at you. Peace Pilgrim

1 0

2025

2024

2023

2022