- Corpora - ELRA lists

Employment opportunities: Research Assistant Position for corpora and data mining at Hong Kong Baptist University
by janice pan 21 Aug '23

21 Aug '23

*Hong Kong Baptist University* *Research Assistant (Full-time) * We invite applications for a full-time Research Assistant position for the project "Re-testing the Universals: Mining Interpreting Data through Re-Engineered Mega-Size Corpora". This pioneering project is at the intersection of corpus-based interpreting studies, natural language processing, and text mining. The project involves working with three significant corpora, namely the Chinese/English Political Interpreting Corpus (CEPIC), the European Parliament Interpreting Corpus (EPIC), and the European Parliament Interpreting Corpus Ghent (EPICG). The successful candidate will be involved in the analysis of these corpora in languages including Chinese, English, Italian, French, Dutch. *Responsibilities:* - Primary focus on data cleaning, preparation, and initial analysis of the CEPIC, EPIC, and EPICG corpora across multiple languages - Collaborative work on hypothesis and methodology development - Application of natural language processing techniques and text mining to analyse and draw insights from the corpora *Requirements:* - A bachelor's degree or above in a related field such as Translation Studies, Linguistics, Data Science, or Computational Linguistics. Recent university graduates are also encouraged to apply. - Proficiency in text mining and programming languages such as Python and R, or tools such as SPSS, for data processing and analysis. - Basic competence in at least two of the languages used in the project: Chinese, English, Italian, French and Dutch. - Prior experience or familiarity with the CEPIC, EPIC, and EPICG corpora will be a significant advantage. - A strong background in computational linguistics or a related field is preferred. *Benefits and Training:* The Research Assistant will be trained and supervised by the project investigators. The initial appointment will be made on a fixed-term contract for one year, with the possibility of further extension. Salary will be competitive, in the range of 20,000-24,000 HKD per month, commensurate with qualifications and experience. For further information, please contact Dr. Janice Pan at janicepan(a)hkbu.edu.hk. *Application Procedures:* Applicants are invited to submit their CV, detailing their relevant experience to the project to Dr. Janice Pan at janicepan(a)hkbu.edu.hk. Review of applications is ongoing until the position is filled.

1 0

CFP: Special Issue on Explainability of NLP models, TAL journal
by Marianna Apidianaki 18 Aug '23

18 Aug '23

Call for Papers Special Issue of the TAL Journal: Explainability of NLP models https://tal-64-3.sciencesconf.org/ ** Deadline for submission: 15 October 2023 ** ** Guest Editors ** Marianna Apidianaki, University of Pennsylvania Guillaume Wisniewski, Université Paris Cité, LLF Neural models build language representations without direct supervision, and have contributed to the important advances of language and speech processing systems in recent years. The use of the representations built by pre-trained language models (such as BERT) allows the development of systems for many languages and domains. However, model decisions are not interpretable. It is often not possible to know why a system makes a specific decision, and the reasons behind the good performance of state-of-the-art models remain mostly unknown. An increasing number of studies attempt to provide answers to these questions by addressing the explicability of NLP systems, and exploring questions related to: • model analysis and interpretability, aimed at identifying the information encoded in neural representations [1] ; • the development of methods aimed at describing and justifying the reasoning steps that allowed systems to reach a specific answer, for example with chain-of-thought prompting (model explicability) [2] ; • the development of "safe" systems, capable of self-justification (model accountability). This special issue of the TAL journal aims to address explanation and analysis methods that have been proposed for NLP systems, as well as our current understanding of the linguistic capabilities of neural models and their limitations. We welcome submissions on the following topics (non-exclusively): • Methods aimed at explaining the decisions of neural networks (identification of important elements, explanation in textual format, …), and specifically methods that allow to establish causality relations between a prediction and the input (or part of the input); • Probing methods aimed at identifying linguistic and world knowledge encoded in neural representations; • Methods that allow to distinguish information that is encoded by neural networks, from information that is actually used by the models for making a decision (correlation versus causality); • Methods inspired by analysis methodology from related domains (experimental linguistics, computer vision, psychology, …); • Bias identification in neural language models; • Prompting methods to generate explanations with text [2] or structured representations [4]; • Study of artificial languages or of linguistically motivated examples; • Evaluation of explanation methods and especially methods aimed at evaluating the faithfulness of explanations. [1] Interpretability and Analysis in Neural NLP (Belinkov et al., ACL 2020) [2] Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (Wei et al., NeurIPS 2022) [3] Towards Faithful Model Explanation in NLP: A Survey (Lyu et al., arxiv 2023) [4] Causal Reasoning of Entities and Events in Procedural Texts (Zhang et al., EACL Findings 2023) ** Important dates ** • Submission deadline: 15 October 2023 • Notification to the authors after first review: December 2023 • Notification to the authors after second review: February 2024 • Publication : April 2024 ** Submission format ** The length of the papers must be between 20 and 25 pages. 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 on this platform: https://tal-64-3.sciencesconf.org/ To do so, authors will need to first create an account by clicking on "Create account" (Créer un compte) next to the “Login" (Connexion) button at the top of this page. To submit a paper, authors can connect to their account and upload their submission in "My Space" > "My submissions”. The articles can be written in English or in French. Papers in English need to include a title and abstract in French. The TAL journal has a double-blind review process. It is necessary to anonymize the article and the name of the file, and to avoid self-references. Each article is evaluated by three reviewers, two external reviewers and a member of the editorial board of the journal TAL. ** TAL Journal ** TAL (Traitement Automatique des Langues / Natural Language Processing) is an international journal published by ATALA (French Association for Natural Language Processing, http://www.atala.org) since 1959 with the support of CNRS (National Centre for Scientific Research). It has moved to an electronic mode of publication, with printing on demand. The TAL journal is open-access. Paper submission, publication and access are free of charge. Papers published in the TAL journal will be made available on the ATALA website and on ACL Anthology.

1 0

Registration for AIRiAL 2023 is now open
by Voss, Erik 17 Aug '23

17 Aug '23

Artificial Intelligence Research in Applied Linguistics (AIRiAL) 2023 Conference Dear Colleagues, Registration for *AIRiAL 2023* is now open. The conference will be held in person on Friday and Saturday, September 29-30, 2023 at Teachers College in New York City. A banquet has been planned for Saturday evening. The early registration discount will be available through *Thursday, August 24, 2023.* AIRiAL Conference site <https://sites.google.com/tc.columbia.edu/airialconference/home?pli=1> *AIRiAL 2023 r*egistration page <https://sites.google.com/tc.columbia.edu/airialconference/airial-2023/regis…> We are looking forward to seeing you in New York City! Best, The AIRiAL 2023 Organizing Committee -- Erik Voss, Ph.D. Assistant Professor, Applied Linguistics & TESOL program Language & Technology Specialization Department of Arts & Humanities Teachers College, Columbia University TC Faculty Profile <https://www.tc.columbia.edu/faculty/ev2449/>, Linkedin Profile <https://www.linkedin.com/in/erik-voss-ph-d-941a3ab9>, Google Scholar <https://scholar.google.com/citations?user=FMnVdjcAAAAJ&hl=en> ALTESOL Language & Technology Research Group <https://sites.google.com/tc.columbia.edu/al-tesol-language-technology/home> AIRiAL 2023 Conference <https://sites.google.com/tc.columbia.edu/airialconference/home> *Latest Publications* Voss, E. (2022). Argument-based validation in the time of the COVID-19 pandemic <https://www.taylorfrancis.com/chapters/edit/10.4324/9781003221463-6/argumen…>. (Ch. 5) Routledge. Voss, E. (2023). Proctoring remote language assessments <https://www.routledge.com/Fundamental-Considerations-in-Technology-Mediated…>. (Ch. 12) Routledge.

1 0

Two-year post doc position at the University of Genoa (Italy) - Deadline August 31st 2023
by mauro.giacomini＠unige.it 17 Aug '23

17 Aug '23

Dear all, we are looking for a Research Assistant to work on the project Digital Health Solutions in Community Medicine (DHEAL-COM), financed by the NextGeneration EU program through the Italian Health Ministry. The position is for two years. The successful candidate will be based in the Health Informatic Section of the Bioengineering Laboratory at the Department of Informatics, Bioengineering, Robotics and System Engineering (DIBRIS - https://dibris.unige.it/) of the University of Genova - Italy. The overall objective of the Digital Health Solutions in Community Medicine project (DHEAL-COM) Project is the strengthening and development of a Life Science Hub at national level (DHEAL-COM HUB) for the co-creation of digital health services where the person’s needs are the starting point. The DHEAL-COM HUB will act as a central integrator, to bring together and coordinate three satellite national Competence Centre Nodes that will focus on: Digital Health Model/Services, Digital Health Technologies, and Digital Health Usability/Accessibility. Specifically, the title of the call is: Electronic Clinical Data Management: Developing a machine learning-based system for extracting meaningful information from clinical records to fuel a growing data lake. The goal of the two-year research is to collect high-quality data to support the construction and growth of the DHEAL-COM data lake. For this purpose, it will be necessary to automatically remove or replace information that directly identifies the patient, in order to process large amounts of data. Secondly, it will be necessary to extract fundamental information for diagnosis and treatment from data, both structured and unstructured, and codify them according to international standards (eg: ICD11) to allow sharing of the same across several national and international centres. These activities require a strong knowledge of NLP techniques which will have to be widely applied to achieve the purposes of the call. The official call and is available here: https://concorsi.unige.it/home/procedure/4125 Deadline for submission is: August 31st Constraint: the person must have a PhD or obtain it during the first year of the contract Kind regards Mauro

1 0

Second Call for Papers: The 4th Workshop on Evaluation and Comparison of NLP Systems (Eval4NLP 2023)
by Juri Opitz 15 Aug '23

15 Aug '23

Dear colleagues, the 4th Workshop on Evaluation and Comparison for NLP systems (Eval4NLP), co-located at the 2023 Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (AACL 2023), invites the submission of long and short papers, with a theoretical or experimental nature, describing recent advances in system evaluation and comparison in NLP. ****_New_****: This year's edition of the Eval4NLP workshop puts a focus on the evaluation of and through large language models (LLMs). Notably, the workshop will feature a shared task on LLM evaluation and specifically encourages the submission of LLM evaluation focused papers. Other submissions that fit the general scope of Eval4NLP are of course also welcome. See below for more details. ** Important Dates ** All deadlines are 11.59 pm UTC -12h (“Anywhere on Earth”). - Direct submission to Eval4NLP deadline: August 25 - Submission of pre-reviewed papers to Eval4NLP (see below for details) : September 25 - Notification of acceptance: October 2 - Camera-ready papers due: October 10 - Workshop day: November 1 Please see the Call for Papers for more details: https://eval4nlp.github.io/2023/cfp.html. ** Shared Task ** This year’s version will come with a shared task on explainable evaluation of generated language (MT and summarization) with a focus on LLM prompts. Please find more information on the shared task page: https://eval4nlp.github.io/2023/shared-task.html. ** Topics ** Designing evaluation metrics: Proposing and/or analyzing metrics with desirable properties, e.g., high correlations with human judgments, strong in distinguishing high-quality outputs from mediocre and low-quality outputs, robust across lengths of input and output sequences, efficient to run, etc.; Reference-free evaluation metrics, which only require source text(s) and system predictions; Cross-domain metrics, which can reliably and robustly measure the quality of system outputs from heterogeneous modalities (e.g., image and speech), different genres (e.g., newspapers, Wikipedia articles and scientific papers) and different languages; Cost-effective methods for eliciting high-quality manual annotations; and Methods and metrics for evaluating interpretability and explanations of NLP models. Creating adequate evaluation data: Proposing new datasets or analyzing existing ones by studying their coverage and diversity, e.g., size of the corpus, covered phenomena, representativeness of samples, distribution of sample types, variability among data sources, eras, and genres; and Quality of annotations, e.g., consistency of annotations, inter-rater agreement, and bias check. Reporting correct results: Ensuring and reporting statistics for the trustworthiness of results, e.g., via appropriate significance tests, and reporting of score distributions rather than single-point estimates, to avoid chance findings; reproducibility of experiments, e.g., quantifying the reproducibility of papers and issuing reproducibility guidelines; and Comprehensive and unbiased error analyses and case studies, avoiding cherry-picking and sampling bias. ** Submission Guidelines ** The workshop welcomes two types of submission -- long and short papers. Long papers may consist of up to 8 pages of content, plus unlimited pages of references. Short papers may consist of up to 4 pages of content, plus unlimited pages of references. Please follow the ACL ARR formatting requirements, using the official templates: https://github.com/acl-org/acl-style-files. Final versions of both submission types will be given one additional page of content for addressing reviewers’ comments. The accepted papers will appear in the workshop proceedings. The review process is double-blind. Therefore, no author information should be included in the papers and the (optional) supplementary materials. Self-references that reveal the author's identity must be avoided. Papers that do not conform to these requirements will be rejected without review. ** The submission sites on Openreview ** Standard submissions: https://openreview.net/group?id=aclweb.org/AACL-IJCNLP/2023/Workshop/Eval4N… Pre-reviewed submissions: https://openreview.net/group?id=aclweb.org/AACL-IJCNLP/2023/Workshop/Eval4N… See below for more information on the two submission modes. ** Two submission modes: standard and pre-reviewed ** Eval4NLP features two modes of submissions. Standard submissions: We invite the submission of papers that will receive up to three double-blind reviews from the Eval4NLP committee, and a final verdict from the workshop chairs. Pre-reviewed: To a later deadline, we invite unpublished papers that have already been reviewed, either through ACL ARR, or recent AACL/EACL/ACL/EMNLP/COLING venues (these papers will not receive new reviews but will be judged together with their reviews via a meta-review; authors are invited to attach a note with comments on the reviews and describe possible revisions). Final verdicts will be either accept, reject, or conditional accept, i.e., the paper is only accepted provided that specific (meta-)reviewer requirements have been met. Please also note the multiple submission policy. ** Optional Supplementary Materials ** Authors are allowed to submit (optional) supplementary materials (e.g., appendices, software, and data) to improve the reproducibility of results and/or to provide additional information that does not fit in the paper. All of the supplementary materials must be zipped into one single file (.tgz or .zip) and submitted via Openreview together with the paper. However, because supplementary materials are completely optional, reviewers may or may not review or even download them. So, the submitted paper should be fully self-contained. ** Preprints ** Papers uploaded to preprint servers (e.g., ArXiv) can be submitted to the workshop. There is no deadline concerning when the papers were made publicly available. However, the version submitted to Eval4NLP must be anonymized, and we ask the authors not to update the preprints or advertise them on social media while they are under review at Eval4NLP. ** Multiple Submission Policy ** Eval4NLP allows authors to submit a paper that is under review in another venue (journal, conference, or workshop) or to be submitted elsewhere during the Eval4NLP review period. However, the authors need to withdraw the paper from all other venues if they get accepted and want to publish in Eval4NLP. Note that AACL and ARR do not allow double submissions. Hence, papers submitted both to the main conference and AACL workshops (including Eval4NLP) will violate the multiple submission policy of the main conference. If authors would like to submit a paper under review by AACL to the Eval4NLP workshop, they need to withdraw their paper from AACL and submit it to our workshop before the workshop submission deadline. ** Best Paper Awards ** We will optionally award prizes to the best paper submissions (subject to availability; more details to come soon). Both long and short submissions will be eligible for prizes. ** Presenting Published Papers ** If you want to present a paper which has been published recently elsewhere (such as other top-tier AI conferences) at our workshop, you may send the details of your paper (Paper title, authors, publication venue, abstract, and a link to download the paper) directly to eval4nlp(a)gmail.com. We will select a few high-quality and relevant papers to present at Eval4NLP. This allows such papers to gain more visibility from the workshop audience and increases the variety of the workshop program. Note that the chosen papers are considered as non-archival here and will not be included in the workshop proceedings. ------------------------------------------------- Best wishes, Eval4NLP organizers Website: https://eval4nlp.github.io/2023/index.html Email: eval4nlp(a)gmail.com

2 3

Postdoc and PhD position in NLP/CL/CSS at GESIS (Cologne) - deadline: Sept 5/6
by Gabriella Lapesa 15 Aug '23

15 Aug '23

Postdoc and PhD position in NLP/CL/CSS at GESIS (Cologne) The newly established Data Science Methods team led by Gabriella Lapesa [2,3] (Leibnitz Institute for Social Sciences GESIS, Cologne [1], Computational Social Science department [4]) has two positions available from November 2023: - one postdoctoral researcher (100%, 4 years, with possibility of tenure) - one doctoral researcher (75%, 4 years). The PhD project will be pursued at the Heinrich Heine University of Düsseldorf (where Gabriella Lapesa is a junior professor in Responsible Data Science and Machine Learning). ** The team ** The Data Science Methods team will contribute to build and mantain the GESIS infrastructure for Computational Social Science (CSS) research by developing novel methods and making them available, documented, and accessible through the GESIS services. The team will focus on fostering the interaction between Natural Language Processing and Social Science by developing solutions that allow for the integration of multiple information sources (e.g., different textual sources for the same debate; socio-demographic features of speakers and audiences; integration of textual and multimodal data) and address recent challenges in NLP (modeling subjective phenomena; low-resource scenarios; identifying and mitigating bias). The team will tackle research questions at the interface between computational argumentation and CSS, and target political communication from a very broad perspective involving different types of actors (citizens, politicians, parties) and discourse contexts (e.g., online discussions vs. newspapers). From a methodological perspective, at the core of the team's research agenda will be the “learning from disagreements” challenge, as machine learning approaches which rely on gold standards which average annotators’ perspectives are particularly unsuitable for the highly subjective phenomena tackled in CSS research (e.g., persuasion in online discussions; harmful communication online; polarization). ** How to apply ** The official job announcement with more details about the requirements/tasks and the application procedure can be found at the following links: Postdoctoral researcher (deadline: September 5th): https://www.hidden-professionals.de//HPv3.Jobs/Gesis//stellenangebot/33073/1 Doctoral researcher (deadline: September 6th): https://www.hidden-professionals.de//HPv3.Jobs/Gesis//stellenangebot/33084/1 [1] https://www.gesis.org/en/home [2] https://www.gesis.org/institut/mitarbeitendenverzeichnis/person/Gabriella.L… [3] https://www.ims.uni-stuttgart.de/institut/team/Lapesa/ [4] https://www.gesis.org/en/institute/departments/computational-social-science

2 3

August 2023 Newsletter - LDC
by Penn LDC 15 Aug '23

15 Aug '23

In this newsletter: LDC at Interspeech 2023 LDC releases speech activity detector Fall 2023 LDC Data Scholarship Program New publications: 2019 OpenSAT Public Safety Communications Simulation<https://catalog.ldc.upenn.edu/LDC2023S06> Samrómur Queries Icelandic Speech 1.0<https://catalog.ldc.upenn.edu/LDC2023S05> ________________________________ LDC at Interspeech 2023 LDC is happy to be back in person as an exhibitor and longtime supporter of Interspeech, taking place this year August 20-24 in Dublin, Ireland. Stop by Stand A2 to say hello and learn about the latest developments at the Consortium. LDC is also delighted to once again be a silver sponsor for the Young Female Researchers in Speech Workshop<https://sites.google.com/view/yfrsw-2023> and to provide data in support of the CHiME-7 challenge<https://www.chimechallenge.org/current/workshop/index> satellite workshop and the MERLIon CCS Challenge<https://sites.google.com/view/merlion-ccs-challenge>. LDC will post conference updates via our social media platforms. We look forward to seeing you in Dublin! LDC releases speech activity detector LDC announces the release of the LDC Broad Phonetic Class Speech Activity Detector. Based on the broad phonetic class recognizer implemented in the HTK Speech Recognition Toolkit<https://htk.eng.cam.ac.uk/>, LDC's speech activity detector model runs the speech signal through a GMM-HMM recognizer to identify five broad phonetic classes: vowel, stops/affricate, fricative, nasal, and glide/liquid. The LDC Broad Phonetic Class Speech Activity Detector is available at no cost on github<https://github.com/Linguistic-Data-Consortium/ldc-bpcsad> under a GPL v3 license<https://www.gnu.org/licenses/gpl-3.0.en.html>. Fall 2023 LDC Data Scholarship Program Student applications for the Fall 2023 LDC Data Scholarship program are being accepted now through September 15, 2023. This program provides eligible students with no-cost access to LDC data. Students must complete an application consisting of a data use proposal and letter of support from their advisor. For application requirements and program rules, visit the LDC Data Scholarships page<https://www.ldc.upenn.edu/language-resources/data/data-scholarships> ________________________________ New publications: 2019 OpenSAT Public Safety Communications Simulation<https://catalog.ldc.upenn.edu/LDC2023S06> contains 141 hours of English speech recordings and transcripts used in the NIST Open Speech Analytic Technologies (OpenSAT<https://www.nist.gov/itl/iad/mig/opensat>) 2019 evaluation's automatic speech recognition, speech activity detection, and keyword search tasks. The data is part of the SAFE-T (Speech Analysis For Emergency Response Technology) corpus created by LDC which is comprised of speakers engaged in a collaborative problem-solving activity representative of public safety communications in terms of speech content, noise types, and noise levels. US English speakers played the board game Flash Point Fire Rescue. Background noise was played through a participant's headset during the recording session. Recording sessions consisted of 2 30-minute games. The corpus is divided into training, development, and evaluation data. 2023 members can access this corpus through their LDC accounts. Non-members may license this data for a fee. * Samrómur Queries Icelandic Speech 1.0<https://catalog.ldc.upenn.edu/LDC2023S05> was developed by the Language and Voice Lab, Reykjavik University<https://lvl.ru.is/> in cooperation with Almannarómur, Center for Language Technology<https://almannaromur.is/>. The corpus contains 20 hours of Icelandic prompted queries from 3,809 speakers representing 17,475 utterances. Speech data was collected between October 2019 and December 2021 using the Samrómur website<https://samromur.is> which displayed prompts to participants. The prompts were mainly from The Icelandic Gigaword Corpus<http://clarin.is/en/resources/gigaword>, which includes text from novels, news, plays, and from a list of location names in Iceland. Additional prompts were taken from the Icelandic Web of Science<https://www.visindavefur.is/> and others were created by combining a name followed by a question or a demand. Prompts and speaker metadata are included in the corpus. 2023 members can access this corpus through their LDC accounts provided they have submitted a completed copy of the special license agreement. Non-members may license this data for a fee. To unsubscribe from this newsletter, log in to your LDC account<https://catalog.ldc.upenn.edu/login> and uncheck the box next to "Receive Newsletter" under Account Options or contact LDC for assistance. Linguistic Data Consortium<ldc.upenn.edu> University of Pennsylvania T: +1-215-573-1275 E: ldc(a)ldc.upenn.edu<mailto:ldc@ldc.upenn.edu> M: 3600 Market St. Suite 810 Philadelphia, PA 19104

1 0

Conference on Harmful Online Communication (CHOC2023), 16-17 Nov, Cologne Germany & online
by Sen, Indira 15 Aug '23

15 Aug '23

Dear colleagues, We (Christina Dahn, Pascal Siegers, Katrin Weller, and I) are hosting the Conference on Harmful Online Communication (CHOC2023) later this year which might also interest NLP researchers working on abusive language detection. The conference will take place in Cologne, Germany, and online on November 16-17, and is generously funded by the Thyssen Foundation. CHOC 2023 aims to bring together researchers and practitioners working on detecting harmful language from a variety of disciplinary lenses. There are also some options for joining with a poster presentation or as participants. See https://www.gesis.org/en/research/conferences/gesis-conferences/conference-… for additional information. We would appreciate it if you could also share this with other people who might be interested. Best wishes, Indira Sen

1 0

BCS Information Retrieval SG - Call for Tutorials: Search Solutions 2023
by Udo Kruschwitz 15 Aug '23

15 Aug '23

Search Solutions is the BCS Information Retrieval Specialist Group’s annual event focused on practitioner issues in the arena of search and information retrieval. Search Solutions consists of two parts: a tutorial day and a conference day. We invite tutorial proposals which focus on any area of the practical application of search technologies to real world problems, for the tutorial day due to take place on 21st Nov 2023 before the conference day on 22nd Nov 2023. Tutorials in previous years have included: designing usability for search, multimedia information retrieval, evaluation, pattern search, city search in SmartCities, text analysis, introduction to natural language processing and introduction to reinforcement learning, Apache Solr and open-source technologies, etc. The details of the previous tutorials can be found here: <>https://www.bcs.org/membership-and-registrations/member-communities/informa… <https://www.bcs.org/membership-and-registrations/member-communities/informa…>. Tutorials Proposals for both full day (5-6 hours including breaks and lunch) and half day (2-3 hours including breaks) tutorials are invited. The tutorials will take place on Tuesday 21st November 2023 at the BCS offices in London and/or online depending on the situation near the time. We encourage in person tutorials at the BCS offices if possible. Proposal submission Tutorial proposals should be submitted to the tutorial chair (h.liu(a)soton.ac.uk <mailto:h.liu@soton.ac.uk>) by midnight Monday 11th Sep 2023, using the following template: Name of presenter(s): please list the names and affiliations of presenter(s). Title: title of the tutorial. Contact details: email and snail mail address, phone numbers etc. Type of tutorial: half day or full day. Delivery format: Online only or in person only or could be both Tutorial Abstract: for publicity. Target audience: please outline the practitioner audience to be addressed. Learning outcomes: what would the practitioners gain from attending this tutorial? Tutorial schedule and description: provide a draft schedule and detailed description of each of the items. Tutorial logistics/materials: required media and formats for tutorial. What will be provided to attendees (e.g. slides). Bio of presenter(s): including track record of presenting tutorials, lecturing experience etc. (200/300 words) Selection Procedure All tutorial proposals will be reviewed by the tutorial chair and approved by the organising committee. The selection criteria will focus on the quality of the tutorial content and the appropriateness of it to the main theme of search solutions. Contact Tutorial Chair: Dr Haiming Liu, h.liu(a)soton.ac.uk <mailto:h.liu@soton.ac.uk> Dr Haiming Liu, PhD, PgCAP, SFHEA Associate Professor @ Web and Internet Science (WAIS) Research Group Director of Centre for Machine Intelligence (CMI) School of Electronics and Computer Science (ECS) Faculty of Engineering and Physical Science University of Southampton Highfield, Southampton, SO17 1BJ Email: h.liu(a)soton.ac.uk <mailto:h.liu@soton.ac.uk>

1 0

Re: math and language/text data (continued from thread "Re: Re: Any literature about tensors-based corpora NLP research with actual examples (and homework ;-)) you would suggest? ...")
by Anil Singh 14 Aug '23

14 Aug '23

Re "Wanderers, Kings, Merchants: The Story of India through Its Languages by Peggy Mohan": look forward to reading it. (History or historical events aside, does it have to do with how language(s) became an indicator of power/privilege/status? If humankind has anything in common, this could be one general observation...) It does to some extent, particularly with regard to the history of languages and related things in India. Re "If people are used to writing in their own language on computers, then that language is more likely to survive": I don't disagree, but --- [Forewarning: one might not like my reply to the following, but I ask for it to be interpreted with a scientific mindset rather, not with emotions and sentiments related to language identity and particular cultural practices.] "Languages" (as in, particular language varieties, not "language" as in language-at-large) come and go, born and die (and, sometimes, come back), like trends, styles, cultures ("culture" as in a particular way of living, a set of habits...). Esp. for users of varieties that have undergone oppression/suppression, I understand that there is or can be much meaning to many users in having the varieties be alive or in use. It is important to the users. It is a symbol of their existence. Yes, that is important. But having witnessed how language has been abused (e.g. with research greed by some CL/NLPers), I sometimes think one might have gone too far with how much identity one attaches to any particular language. Coming trom the background that I do, your statement above seems very similar to saying that we might have gone too far with political correctness or about opposing racism/misogyny/etc. Like most Europeans or Americans, you seem to have no (or very little) idea about the toll that discrimination -- even unintended discrimination -- takes on a very large part of the human population. In early 2009, I had written a rant in a blog post and the title of the post was "English is Language Independent". I had made roughly the same points which the now famous Bender Rule makes. I have been writing about it, although that blog is now defunct. I did not, however, make any proposal, as the Bender Rule does. I just pointed out the problem, so I am in no way undermining the importance of that. And one anonymous comment on this blog post was this: "Why don't you work on a good project. I don't see the prejudice persisting for long once you do that." It is like saying about gender or racism: "Why don't you make accomplishments equal to us? Once you do that, I don't see the prejudice persisting for long." Not to be picky here, but "I have heard some native speakers *users *of some "Dravidian languages" say that there are some (I guess minor) problems with Unicode for their languages": using the terms "native" and "speakers" to refer to "users" (or as I sometimes use among knowers of language: "languagers") has been an unhealthy baggage from our past practices in the language space. I can't comment on the issue of "Dravidian languages" and Unicode, but it seems one might need more info/details on the complaints to act further. Well, yes, the word "native" has indeed a very dark history. How could I not know it? But, the manner in which linguists use the term "native speaker" is very different. At least I think so. Note that I am only mentioning linguists here, not "CL/NLPer". By the way, I thought I had coined the term "NLPer" in my blog long ago, but I may be wrong about that. I am pretty sure one can find out these details in some online forum or some academic publication etc. I have not so far done that, but it is a good idea to do that. I will try. Re "psycholinguistic validity from computational validity": in (cognitive/psycholinguistic) modeling, there is / can be not much difference between the two. (When one has enough experiences with modeling or with language/psycholinguistic phenomena, it's not hard to to see that results from computational modeling could also hold elsewhere. The art then is to be able to connect the two "realms". But then again, it depends on the claims, of course.) Yes, of course, it depends on the claim. The two realms can definitely be connected. We don't disagree about that. In fact, I think, we don't really disagree about that many things it seems. Even so, isn't it possible to implement the same thing in many different ways when it comes to computation? That is not the case with the brain/mind, of course. Here, I am again making a distinction between computation and mathematics. Perhaps you don't agree with that? In that case, perhaps we mean different things by the term "computation". Re science and engineering: I am not sure if engineering has to be "just about" heuristics or short cuts. There is good engineering and there is bad engineering, for the sake of my arguments here. In the context of ML with language/textual data, one ought to be careful with "computing based on values of surface elements/strings". Much of what the CL/NLP community/communities have been doing for the past few decades has been "computing based on values of surface elements/strings". This practice deserves serious re-evaluation (there are lots of grey areas here and opportunities to compare processing across finer granularities (without all the preprocessing hacks/"heuristics"/"engineering") for various tasks and data types/formats, without "words", "sentences", "linguistic structure(s)", "grammar" et al.). I don't think of it as "it's engineering!", but some bad practices/culture having been promoted as such and normalized (for a couple of decades?). Good engineering can also be fine, thoughtful, and robust. I completely agree. But sometimes I do work on things which, theoretically, seem ridiculous to me, but they may be practically useful. At least half (perhaps more) of my motive to work on language processing is to address somehow, to any extent, the issue of linguistic empowerment. I am prepared to compromise theoretically for that purpose. Re "I don't think there is anything wrong with what you call grammar hacking from the engineering point of view": I do (think there is something wrong), because: i. "all grammars leak" (from Sapir, also in Manning and Schütze (1999)); I know. I tell this to students every time I teach NLP or any related subject. ii. "words" (whatever they are) are too coarse-grained for computing. I already agreed to that, but if they help in my benign motives, I am prepared to use them. Re "it [language(s)] is still likely to have an 'organic' structure": couldn't that structure (one not associated with "words"/"sentences"/"grammar") be one from math or computing? Or one that is a by-product of a combination of these factors? It certainly could. Some CL/NLPers have made various claims concerning "structures" in the past, borrowing the concepts from "linguistic structure(s)", from "grammar". There was a lot of chiming along, many often have neglected the fact that grammar could effect the impression of "structures" through "words" etc. or that it all in turn patterns some of our thoughts/judgments sub-/unconsciously. And the loop goes on. (See also: See also: https://twitter.com/adawan919/status/1532335891448057858) Well, if you prefer the term "patterns" to "grammar" or "structure", I am completely fine with that. As I said earlier, I am moving towards the language games view of language, even for this exchange. We can't avoid that if we are talking in a human/natural language. The only way to avoid that, if there is a way, is to use only mathematical notation, but I don't think we have reached that stage so far with the study of language. Re "in English "John loves Mary" is in fact a very different thing than "Mary loves John"": one has to re-evaluate to which extent this matters in whichever form of computing/computation one is engaged in and how often this "canonical form" that you are implicitly referring to really occurs in data as well as how this actually surfaces in data. One should look at the data in front of one, not the framework/theory in one's mind. (I believe in achieving better designs/systems through testing from both a data-centric as well as an algorithm-centric perspective. Hardware counts too!) I mostly agree, but I am not sure whether you are saying that "John loves Mary" may not perhaps be different from "Mary loves John"? Re " it is unfair to blame Linguistics for that": My focus in "[t]he "non-native speakers of X" has been a plague in Linguistics" was on the "native" part. That has been my understanding, at least to a great extent "nativeness" was so promoted/reinforced, esp. within the school of generative Linguistics in the 2nd half of the 20th century, when it comes to "linguistic judgments". I thought the propagation stemmed from there. Who/What else do you think started it? I think the word "native" was used in a derogatory/condescending way throughout the English speaking world, even before the "birth of Modern Linguistics". It was, in fact, the more polite word. One other common word was "savage". I remember being shocked to find (very long ago) in Jane Eyre the phrase "savages living on the banks of the Ganges". Savages? On the bank of the Ganges? But these usages are much older (than "Modern Linguistics"). The matter of "linguistic judgements" or "grammaticality" is very different from that, regardless of what one's opinions are about the existence of grammar. All in all, your replies remind me of many of the reviewers' responses "typical" of (i.e. I often got from) the *CL circle (of those who remained in the past decade or so). I don't know about that. I thought I had very unconventional views of NLP, but I could be wrong about that, at least relatively. If I may guess: i. you don't have an academic background in Linguistics (esp. general Linguistics, That is true. I have learned about language(s) mostly on my own. So, if you want me to show my degree in Linguistics, I have none, except a PhD in Computational Linguistics. I was the second person to get a PhD specifically in CL in India. note that there is a difference between linguistics of particular languages and that in a more general/theoretical manner (not about (p-)language grammatical particularities), Of course there is. What makes you think I don't know that? The fact is, my knowledge of "(p)-language" is relatively very limited. Even my knowledge of the syntax of Hindi (my "mother tongue"), in a formal sense, is very limited. I mostly know about language in general. ii. you learn about language(s) through mostly non-academic books or through your own language experience(s) (which counts too, I am not invalidating it/them here), I have no idea what makes you say that. Am I supposed to list the Linguistics books I have read, in addition to showing my Linguistics degree? (Sorry if that sounds bitter, but it has happened in the past, not literally, but effectively). I can only say -- and it is strange to even have to say this -- that I definitely know more about language in general and linguistic theory than -- at least -- most graduates and postgraduates (including PhDs) of Linguistics in India. iii. you never had phonetics and phonology, nor In my replies on this thread I have not mentioned anything related to phonetics or phonology. So it must be from somewhere else that you have this impression(?). Is it from some of the papers co-authored by me? I think there are some which could give that impression. Explaining why that could be so, will take this discussion somewhere else. I don't think it is relevant here at all. Is your point that I don't know about phonetics or phonology? If it is, I would prefer not to answer that. iv. do you realize how you can practice without "words" I do. But I am also prepared to use "words" wherever they help. As I wrote earlier, I have worked without "words" sometimes and have argued against them. --- did I get any right? You got -- sort of -- the first and the third, assuming you were asking me to show my Linguistics degree and whether I had formally taken courses on Phonetics and Phonology. I wanted to note this because --- and please do not take offense, it is not meant personally for I respect your expertise and appreciate our exchanges --- for a while, I didn't know where(else) to submit my findings. It wasn't until I got all the rejections with rather shallow comments about language (or language and computing) did I realize the "solidarity" one has built with people with a background similar to yours might have been the driving force of how some computational (general) linguists (as in, "general language scientists" who also do computational work --- there are only a few of us) got chased out of the arena. The "typical" excuses for practices of this "culture" have been "engineering", "useful", "it works" --- but without any/much grounding/interest in good generalizations. One puts excess focus on processing but not on evaluation or interpretation. I think it's time for a "culture" change in this regard. To your other reply below (in triple quotes): Sorry, but I didn't understand what you mean by triple quotes. I could find any triple quotes in your comments. re "language policy": not everything has to be or can be regulated. Policies can help with promoting/reinforcing/rectifying a particular situation/initiative. I agree. Forcing people e.g. to use language in one particular way or to use "one language" only (whatever "language" "means"*)? Again, I agree. However, like most Europeans and Americans, you seem to have no (or very little) idea how people are already being forced to use some language or another. And it is hardly a new phenomenon, but it has become much more serious now due mainly to colonization and all its effects. Do you have any idea how much hundreds of millions of Indians suffer simply from being forced to use English? My primary motive in my whole life, for good reasons, has been to counter linguistic discrimination, mainly due to the imposition of English on any Indian who has any ambition at all. I think that alone makes me sufficiently qualified to "work on language". This is analogous to any other kind of discrimination or prejudice. I do not think that would be a good direction to go. For any regions, we have seen both good/better and bad/worse policies throughout the course of history. One would really have to evaluate the proposed policy in question carefully. I never said anything about forcing people to speak one language. That's why I said I don't know what exactly the connection is with the language policy. But it sure has a very strong connection, because the problem in the first place is due to language policy, written and unwritten. Do you know that there are and have been schools in the world, including India, where students are punished if they are caught speaking in their mother tongue (or first, "native" language). I still remember feeling recognition and the impression it had on me when I read the famous novel about life in Wales, How Green was My Valley. I had just become fluent in English then. Depending on the situation, some may best change things through the economy, some through government support, some through education and/or grassroot-type of initiatives, some a combination of all these and more.... I agree. *I have an answer to this... please wait for my next pub or so. I would love to read it. I am eternally hungry for any fresh look on language in general. Not so much for particular languages or varieties. That is to me, to some extent, boring. Re "it is very much like conservation of ecology or of species. I don't think it (the latter) will be considered unwarranted prescriptivism": see my 2nd response above. The same for my reply to that comment. Also, with language documentation, one can just document data without promoting grammar. (That's probably the less unethical thing one can do with language or language data.) Again I agree. I never said anything about promoting grammar. I don't like to read grammar books. It's painful to me, compared to almost any other topic under the sun, except perhaps finance, commerce, and the intricacies of legal procedures. For the sake of completeness, I should clarify -- as it seems to matter -- that I have "never had Syntax or even Semantics or Pragmatics". I am mostly self-taught, not just in Linguistics, but also in Computer Science and almost everything else. Do you really think it matters in the context of this discussion? Again, for the sake of completeness, I should mention that for decades, I have been reading all kinds of books that had anything to do with language, mostly in general, but also about Hindi or Indian languages, not to mention English. These have included what you call academic books on language in general and about Linguistics. I still keep reading, as I know very well that, being self-taught, I have some gaps in my knowledge of Linguistics and Computer Science. My undergraduate degree was in Mechanical Engineering (from 1990), but I hardly remember anything in that area. I have similarly been reading all kinds of books for decades about computers and Computer Science. I am unable to see how any of this matters in the context of this discussion. By the way, I like the metaphor you use for language: It being like a graphical user interface for the brain. That reminds me of the views of Daniel Dennett about consciousness. He constantly compares elements making up consciousness to graphical user interfaces on computers. Not that I completely agree with him about consciousness, but I still find the metaphor quite good, perhaps as an approximation.

2 5