- Corpora - ELRA lists

SwissText 2025: Final CfP and Submission Deadline Extension
by Tuggener Don (tuge) 13 Mar '25

13 Mar '25

Dear colleagues, partners, and friends This is the final Call for Papers for the SwissText 2025 Conference. !!! Note that we extend the submission deadline to 22.03.2025 (23:59 AOE) !!! If you plan to make a submission, it helps us with the planning if you already submit an abstract and/or a draft now. You can still revise the submission until the submission deadline. We invite you to submit a contribution to the 10th Swiss Text Analytics Conference (SwissText) which takes place on May 14-15, 2025 in Winterthur, Switzerland at the Zurich University of Applied Sciences (ZHAW). SwissText<https://swisstext.org/> is an annual conference that brings together text analytics experts from industry and academia. It is organized by the Swiss Association for Natural Language Processing (SwissNLP) in collaboration with the Zurich University of Applied Sciences (ZHAW). This edition of SwissText will be special, since we celebrate its 10th anniversary! For the occasion, we introduce the Impact Track which replaces and extends the scope of the Applied Track, and we introduce the Corpus Track that offers the opportunity to present Swiss-related NLP datasets. The conference now features the following tracks and submissions (more details in the below sections): * Impact Track (formerly Applied Track): Describe impactful applied and practical NLP projects. Short abstract (approx 0.5-1 page). * Scientific Research: Describe original research. * Regular Track: Full paper (up to 8 pages), short paper (up to 4 pages) * Junior Track: Short paper (up to 4 pages) * Corpus Track (new): Present your Swiss-related NLP dataset. Short paper (up to 4 pages). TRACKS Impact Track (formerly Applied Track) Contribute to the conference by submitting a presentation with a strong focus on practical application and impact. Have you developed an interesting NLP solution that solves a challenging problem in your organization? Did you write an impactful blog article on an NLP-related topic? Have you released a cool new NLP Python library? Is there an important idea presented in a paper you have read or written that you think needs more exposure to a broader audience? The impact track is open to various NLP-related contributions. The selection criteria here is how exciting and relevant the presentation will be for the broad audience. Presentations can be (non-exhaustive list): * Applications * Applied research results * Showcases * Paper discussions (only already published papers; original work can be submitted to the scientific track) * System demonstrations Submission Format: * A PDF document that contains * a short abstract (approx. 0.5-1 page of main content) describing your contribution; * a screenshot or system diagram (optional) * CV of the team (text format, no tables — max. 500 characters) * Description of the intended audience (e.g. developers, decision makers, project managers, researchers; max. 300 characters) * Language: English * Please submit online via <https://easychair.org/conferences/?conf=swisstext2025> EasyChair<https://easychair.org/conferences/?conf=swisstext2025> <https://easychair.org/conferences/?conf=swisstext2025> Scientific Track Contribute to the conference by submitting original papers that illustrate research results, projects, surveying works that describe new concepts, innovative research, systems, or standards in the areas of NLP. (If you have a live demonstration, please submit to the Impact Track instead) Submission Format: * Full Paper of maximum of 8 pages, or Short Paper with maximum of 4 pages (both versions have additional pages for references and appendix) * Use the ACL style<https://github.com/acl-org/acl-style-files> * Language: English * Anonymization: The review process will be double blind; we ask authors to anonymize their submission reasonably. For instance, instead of “We previously showed (Smith, 1991) …” write “Smith previously showed (Smith, 1991) …” * All data is entered online via EasyChair<https://easychair.org/conferences/?conf=swisstext2025> Scientific Junior Track We offer PhD/Master students a unique opportunity to submit ongoing or finished work relevant to the NLP field and receive extensive professional feedback. Several senior researchers in the NLP field in Switzerland have expressed their support for this initiative and will be members of the programme committee. They will provide extensive reviews to the students to help them improve their paper writing skills. Submission Format: * Please use the ACL style<https://github.com/acl-org/acl-style-files> * Language: English * Short Paper with maximum of 4 pages, with additional pages for references and appendix. * Content: ongoing or finished work, relevant to the NLP field. * The first author must be a student registered at a Swiss academic institution. * Anonymization: The review process will be double blind; we ask authors to anonymize their submission reasonably. For instance, instead of “We previously showed (Smith, 1991) …” write “Smith previously showed (Smith, 1991) …” * Submission will be made through EasyChair<https://easychair.org/conferences/?conf=swisstext2025> Corpus Track This special track focuses on the creation, curation, and release of datasets that address challenges and opportunities in NLP in the context of Switzerland. The aim is to provide a platform for researchers, practitioners, and industry professionals to share innovative dataset contributions that support linguistic, cultural, and technical advancements in the Swiss NLP ecosystem. Submissions can describe already published or original work. It might be possible to host your data at Swiss NLP if you are unable to do so yourself. Please get in touch at info(a)swissnlp.org<mailto:info@swissnlp.org> if you are interested in this option. The submission is a “Corpus Description Paper”, which should contain the following: * What and how much data was collected? * How was the data collected and annotated? * Corpus statistics * Quality assessments such as annotator agreements * Where is the corpus available? (please take into account long-term availability) * Under which license(s) is the corpus distributed? * For which use cases can the corpus be used? * Optional: baseline systems and results Submission Format: * Please use the ACL style<https://github.com/acl-org/acl-style-files> * Language: English * Corpus description paper with maximum of 8 pages, with additional pages for references and appendix. * Please use the appendix to describe technical details of the corpus creation, in particular how data was collected, full annotation guidelines, etc. * Content: new/existing NLP datasets within the context of Switzerland. * Submission will be made through EasyChair<https://easychair.org/conferences/?conf=swisstext2025> GENERAL INFORMATION Presentation Modes For all tracks, there are the following presentation modes: * Talk: a classical presentation of approx. 15min + 5min discussion * Poster: you can hang up an A1 poster during the poster session * Demo: you get a booth where you can show your solution as live-demo during the exhibition time (approx 2 hours) During submission you can choose your preferred presentation mode and the SwissText Program Committee will decide which format your submission best fits in. Proceedings We plan to publish the proceedings of the Scientific, Corpus and Junior track in the ACL Anthology. To appear in the proceedings, each contribution has to be presented by at least one author at the conference. Important Dates * Submission deadline: March 22, 2025 (23:59 AOE) * Author notification: April 5, 2025 * Camera-ready version due: April 19, 2025 * Conference: May 14-15, 2025 Venue The conference will take place in Winterthur at the Zurich University of Applied Sciences (ZHAW). Organization * General Chair: Prof. Dr. Mark Cieliebak (ZHAW) * Program Chairs: Dr. Don Tuggener (ZHAW), Manuela Hürlimann (ZHAW) Contact Please direct all questions regarding the submission process to info(a)swisstext.org<mailto:info@swisstext.org>.

1 0

Digital lexicography and lexical computing workshop, Bari, Italy
by Ondřej Matuška 13 Mar '25

13 Mar '25

*<Lexicom/>* a workshop in digital lexicography and lexical computing *Registration open* *Bari, Italy*15 – 19 September 2025 Your 5 days to get up-to-date with the latest developments in *corpus-driven lexicography* and to practice your *corpus building and corpus query skills* with some of the top experts in the field. For the programme, lecturers, invited speakers, fees and registration, visit this website *lexicom.courses <https://lexicom.courses/upcoming-lexicom/>* I hope to meet you in Bari in September! Ondřej *Ondřej Matuška* sketchengine.eu <http://www.sketchengine.eu/> | Facebook <https://www.facebook.com/SketchEngine/> | LinkedIn <https://www.linkedin.com/in/ondrejmatuska> | Twitter <https://twitter.com/SketchEngine>

1 0

Appel : 1st Call for papers: MLP-LLM 2025 Workshop @ CORIA-TALN 2025
by Aman Sinha 13 Mar '25

13 Mar '25

**** French version below **** * Workshop on Medical Language Processing in the era of Large Language Models (MLP-LLM 2025) * Colocated with CORIA-TALN 2025 -- 30 June 2025 , Marseille * Call for Papers * [ https://atilf-umr7118.github.io/MLPLLM2025/ | https://atilf-umr7118.github.io/MLPLLM2025/ ] The advent of large language models (LLMs) has revolutionized natural language processing across various domains, including healthcare. However, the complexities of medical language—marked by specialized terminologies, the use of abbreviations and code ontologies such as ICD, UMLS or SNOMED, implicit contextual dependencies (based on the context, the medication information may be different: temporality, action, certainty, etc.) —pose unique challenges and opportunities. Medical NLP is at critical stakes, given the importance of finding the right diagnosis and treatment for each patient. Moreover, the field of health includes not only the human aspect represented by the practitioner-patient relationship, but also the contact with the biological world (animals, plants, viruses, microbes). This workshop, MLP-LLM, aims to bring together researchers from NLP, medicine, bioNLP and linguistics to explore advancements, limitations, and ethical considerations of using LLMs in medical contexts. Topics of interest include, but are not limited to: * Fine-tuning and adapting LLMs for medical applications and for different languages. * Addressing biases in medical language understanding along with LLM hallucinations. * Proposing evaluation methods to assess the quality of medical NLP tools. * Ensuring transparency, interpretability, and uncertainty awareness in medical AI systems. * Developing domain-specific benchmarks for evaluating LLMs in healthcare. * Developing applications for LLMs in clinical decision support, medical transcription and communication between practitioners and patients. Keynote speaker : Natalia Grabar, Université de Lille We welcome articles that are: * new contributions, * state-of-the-art articles, * work in progress, * short/translated version of a paper accepted at a major conference. Important Dates : * Submission Due: 30 April 2025 * Author Notification: 15 May 2025 * Camera ready: 21 May 2025 * Workshop: 30 June 2025 Submissions are accepted both in English or French. Contact : Workshop Organizers ( [ mailto:mlpllm2025@gmail.com | mlpllm2025(a)gmail.com) ] Ioana Buhnila ( [ mailto:ioana.buhnila@univ-lorraine.fr | ioana.buhnila(a)univ-lorraine.fr ] ) Aman Sinha ( [ mailto:aman.sinha@univ-lorraine.fr | aman.sinha(a)univ-lorraine.fr ] ) --------------------------------------------------------- **** V ersion anglaise ci-dessus **** * L'atelier Traitement du langage médical à l'époque des LLMs (MLP-LLM 2025) * Coloc avec CORIA-TALN 2025 -- 30 juin 2025, Marseille * Appel à communications * [ https://atilf-umr7118.github.io/MLPLLM2025/ | https://atilf-umr7118.github.io/MLPLLM2025/ ] L’avènement des grands modèles de langue (LLMs) a révolutionné le traitement automatique des langues dans divers domaines, y compris la santé. Cependant, la complexité du langage médical - marquée par des terminologies spécialisées, l’utilisation des abréviations et des ontologies médicales de type CIM, UMLS or SNOMED, des dépendances contextuelles implicites (dans un contexte donné, le contenu médical peut changer en fonction de la temporalité, l’événement ou le degré de certitude) - pose des défis et des opportunités uniques. Le TAL médical peut aider les praticiens hospitaliers dans le diagnostic et le traitement des patients. De plus, le domaine de la santé comprend non seulement l’aspect humain représenté par la relation praticien-patient, mais également le contact avec le monde biologique (animaux, plantes, virus, microbes). Cet atelier, MLP-LLM, vise à rassembler des chercheurs en TAL, bioinformatique, médecine et linguistique afin d’explorer les avancées, les limites et les considérations éthiques de l’utilisation des LLMs dans des contextes médicaux. Les sujets d’intérêt incluent, mais ne sont pas limités à: * Affiner et adapter les LLMs pour les applications médicales et pour différentes langues. * Proposer des méthodes d’évaluation adaptées au domaine médical. * Traiter les biais dans la compréhension du langage médical et les hallucinations des LLMs. * Garantir la transparence, l’interprétabilité, le niveau de certitude et la responsabilité dans les systèmes d’IA médicale. * Développer des benchmarks spécifiques au domaine pour évaluer les LLMs dans le domaine de la santé. * Développer des applications des LLMs dans l’aide à la décision clinique, la transcription médicale et la communication entre les praticiens et les patients. Keynote speaker : Natalia Grabar, Université de Lille Les types d'article acceptés sont: * contribution nouvelle, * état de l'art, * travaux en cours, * version courte/traduite d'un article accepté dans une grande conférence. Calendrier de l'appel : * Soumission des articles: 30 Avril 2025 * Notification aux auteurs: 15 Mai 2025 * Version finale: 21 Mai 2025 * Atelier : 30 Juin 2025 Les soumissions sont acceptées en anglais ou en français. Comités et contact ----------------- Organizers : Ioana Buhnila, ATILF, CNRS - Université de Lorraine (co-responsable) Aman Sinha, IECL- ATILF - ICANS Strasbourg (co-responsable) Hanbyul Song, ATILF, CNRS - Université de Lorraine Laura Zanella, POSOS Salomé Klein, LiLPa, Université de Strasbourg Joé Laroche, LiLPa, Université de Strasbourg Delphine Charuau, Trinity College Dublin Sam Bigeard, INRIA, Université de Lorraine Comité scientifique : Aurélie Névéol, Université Paris-Saclay, CNRS, LISN Natalia Grabar, Université de Lille Remi Cardon, CENTAL, IL&C, Université Catholique de Louvain Cyril Grouin, LISN, CNRS - Université Paris-Saclay Amalia Todirascu, LiLPa, Université de Strasbourg Lina F. Soualmia, Université Rouen Normandie Patrick Watrin, CENTAL, UCLouvain Adrien Coulet, Inria Paris - Université de Lorraine Benoit Favre, Université d’Aix-Marseille Emmanuel Morin, Université de Nantes Richard Dufour, Université de Nantes Mathieu Constant, Université de Lorraine Timothee Mickus, University of Helsinki Claire Nedellec, Institut National de Recherche Agronomique (INRA) Lisa Raithel, DFKI Ioana Buhnila, ATILF, CNRS - Université de Lorraine Aman Sinha, IECL- ATILF - ICANS Contact : Workshop Organizers ( [ mailto:mlpllm2025@gmail.com | mlpllm2025(a)gmail.com) ] Ioana Buhnila ( [ mailto:ioana.buhnila@univ-lorraine.fr | ioana.buhnila(a)univ-lorraine.fr ] ) Aman Sinha ( [ mailto:aman.sinha@univ-lorraine.fr | aman.sinha(a)univ-lorraine.fr ] )

1 0

CfParticipation: ESSLLI 2025, July 28 - August 8, 2025, Bochum (Germany)
by ESSLLI Publicity 12 Mar '25

12 Mar '25

We invite you to participate in the 36th European Summer School in Logic, Language and Information (ESSLLI), taking place from 28 July - 8 August, 2025 at Ruhr University Bochum, Germany. https://2025.esslli.eu/ * Overview The European Summer School in Logic, Language and Information (ESSLLI) is a yearly recurring event, organized under the auspices of the Association for Logic, Language and Information (FoLLI), and has been running since 1989. The ESSLLI Summer School provides an interdisciplinary setting in which courses and workshops are offered in logic, linguistics and computer science, also from wider scientific, historical, and philosophical perspectives. ESSLLI attracts around 400 participants from Europe, the Middle East, Asia and Africa, as well as from North America and Latin America. ESSLLI has become the main meeting place for young researchers and students in logic, linguistics and computer science to discuss current research and to share knowledge. The event is unique in its interdisciplinary set-up, with no equivalents in Europe. * Programme The ESSLLI Summer School offers an exciting two-week programme, consisting of the following: - Foundational, introductory and advanced courses in three areas: Language and Computation, Logic and Computation, and Logic and Language - Workshops in logic, linguistics and computer science - Student session - Evening lectures - Social activities The full program can be found on the website. * Registration: Registration for attendees is now open. The early-registration deadline is Saturday, 31st May. https://2025.esslli.eu/registration/registration.html * Accommodation Various options for accommodation at special negotiated rates are listed on the website and available first-come, first-served. We have also set up a dedicated Discord server which can be used for coordinating shared apartments, etc.

1 0

Stop notification
by Kayode V Amusan 12 Mar '25

12 Mar '25

stop

1 0

[Call for Papers] Developing Models for Linguistic Research: Training Data Usage in Low-Resource Scenarios
by Cristina Garcia Holgado 12 Mar '25

12 Mar '25

[Call for Papers] Developing Models for Linguistic Research: Training Data Usage in Low-Resource Scenarios 💬 Appel à communications en français : voir fichier joint 💬 Call for papers in english: [ https://llcd2025.sciencesconf.org/data/pages/Developing_models_for_linguist… | https://llcd2025.sciencesconf.org/data/pages/Developing_models_for_linguist… ] Workshop at LLcD 2025 This workshop focuses in approaches and limitations to adapting computational models (such as LLMs) for linguistic research in low-resource scenarios. These include under-resourced, minority, and endangered languages, ancient languages, learner corpora, and language disorder corpora. It seeks to foster interdisciplinary dialogue and collaboration, with the goal of developing best practices, identifying tool limitations, and establishing evaluation guidelines for low-resource linguistic research. 🏙️ Venue and Dates : 1–3 September 2025, Lille, France 📄 Submission Format : Abstract (500 words) 📅 Submission Deadline : 30 March 2025 ➡ Submission Information : [ https://lnkd.in/gKX47Esu | https://llcd2025.sciencesconf.org/resource/page/id/7 ] 🌐 Workshop Languages : English and French We invite contributions from researchers working on corpus construction, transcription, annotation, and analysis, as well as computational linguists and NLP specialists. ❓ For inquiries , please contact Natasha Romanova and Aleksandra Miletić (contact details available in the call).

1 0

1st CFP – Pedagogical Ability Assessment of AI-powered Tutors @ BEA 2025
by Ekaterina Kochmar 12 Mar '25

12 Mar '25

Conversational agents offer promising opportunities for education as they can fulfill various roles (e.g., intelligent tutors and service-oriented assistants) and pursue different objectives (e.g., improving student skills and increasing instructional efficiency), among which serving as an AI tutor is one of the most prevalent tasks. Recent advances in the development of Large Language Models (LLMs) provide our field with promising ways of building AI-based conversational tutors, which can generate human-sounding dialogues on the fly. The key question posed in previous research, however, remains: *How can we test whether state-of-the-art generative models are good AI teachers, capable of replying to a student in an educational dialogue?* In this shared task, we will focus on educational dialogues between a student and a tutor in the mathematical domain grounded in student mistakes or confusion, where the AI tutor aims to remediate such mistakes or confusions, with the goal of evaluating the quality of tutor responses along the key dimensions of tutor’s ability to (1) identify student’s mistake, (2) point to its location, (3) provide the student with relevant pedagogical guidance, that is also (4) actionable. Dialogues used in this shared task include the dialogue contexts from MathDial (Macina et al., 2023) and Bridge (Wang et al., 2024) datasets, including the last utterance from the student containing a mistake, and a set of responses to the last student’s utterance from a range of LLM-based tutors and, where available, human tutors, aimed at mistake remediation and annotated for their quality. **Tracks** This shared task will include five tracks. Participating teams are welcome to take part in any number of tracks. - Track 1 - Mistake Identification: Participants are invited to develop systems to detect whether tutors' responses recognize mistakes in students' solutions. - Track 2 - Mistake Location: Participants are invited to develop systems to assess whether tutors' responses accurately point to genuine mistakes and their locations in the students' responses. - Track 3 - Pedagogical Guidance: Participants are invited to develop systems to evaluate whether tutors' responses offer correct and relevant guidance, such as an explanation, elaboration, hint, examples, and so on. - Track 4 - Actionability: Participants are invited to develop systems to assess whether tutors' feedback is actionable, i.e., it makes it clear what the student should do next. - Track 5 - Guess the tutor identity: Participants are invited to develop systems to identify which tutors the anonymized responses in the test set originated from. **Participant registration** All participants should register using the following link: https://forms.gle/fKJcdvL2kCrPcu8X6 **Important dates** All deadlines are 11:59pm UTC-12 (anywhere on Earth). - March 12, 2025: Development data release - April 9, 2025: Test data release - April 23, 2025: System submissions from teams due - April 30, 2025: Evaluation of the results by the organizers - May 21, 2025: System papers due - May 28, 2025: Paper reviews returned - June 9, 2025: Final camera-ready submissions - July 31 and August 1, 2025: BEA 2025 workshop at ACL **Shared task website**: https://sig-edu.org/sharedtask/2025 **Organizers** - Ekaterina Kochmar (MBZUAI) - Kaushal Kumar Maurya (MBZUAI) - Kseniia Petukhova (MBZUAI) - KV Aditya Srivatsa (MBZUAI) - Justin Vasselli (Nara Institute of Science and Technology) - Anaïs Tack (KU Leuven) **Contact**: bea.sharedtask.2025(a)gmail.com

1 0

2nd Call for Participation: SustainEval 2025 - Understanding Sustainability Reports
by Jakob Prange 12 Mar '25

12 Mar '25

2nd Call for Participation: SustainEval 2025 - Understanding Sustainability Reports We invite system and paper submissions for SustainEval 2025, a GermEval Shared Task co-located with KONVENS 2025 in Hildesheim, Germany, in September 2025. With this shared task, we aim to fuel research on automatic analysis and detection of greenwashing by challenging participants to build systems that categorize excerpts from German-language sustainability reports for (A) content class and (B) statement verifiability rating. We have just published the development data, so you can go ahead and start building your systems! Data repo: https://github.com/SustainEval/sustaineval2025_data Important Dates All submission deadlines are 11:59 p.m. UTC-12:00 "anywhere on Earth."Development Data Release (270 Instances):12th March 2025Training Data Release (1000 Instances):14th April 2025Registration Deadline / Start Evaluation Phase:10th June 2025System/Results Submission Deadline:27th June 2025System Description Paper Submission Deadline:11th July 2025System papers will be reviewed within 10 days after the submission deadlineCamera-ready Deadline:15th Aug 2025Workshop & Final Presentation: 10th Sept 2025 For more information please refer to the shared task homepage: https://sustaineval.github.io/ If you have any questions, please feel free to contact us at sustaineval(a)gmail.com. Best regards and happy holidays, The Shared Task Organizers Jakob Prange, Universität Augsburg (contact for task-specific questions): jakob.prange(a)uni-a.de Charlott Jakob, TU Berlin (contact for organisational questions): c.jakob(a)tu-berlin.de Annemarie Friedrich, Universität Augsburg -- Dr. Jakob Prange (er/he) Akademischer Rat auf Zeit / Research Associate Chair for Natural Language Understanding & Digital Humanities (Prof. Friedrich) Faculty for Applied Informatics, University of Augsburg https://jakpra.github.io/

1 0

KONVENS 2025: Submission now open
by Christian Wartena 12 Mar '25

12 Mar '25

Dear all, We are pleased to inform you that submission for KONVENS 2025 is now open. Papers can be submitted here: https://openreview.net/group?id=GSCL.org/KONVENS/2025 Further Information can be found on the conference website: https://konvens-2025.hs-hannover.de/ Paper Submission Information We invite submissions of original and unpublished work on all aspects of computational linguistics and natural language processing from fundamental inquiries to the practical implementation of natural language resources, components, and systems. We particularly encourage submissions of NLP approaches dedicated to the German language, including survey papers that provide insights into the current state of the art in German language and speech processing. We welcome contributions from both academic and industry professionals. Papers can be submitted either to the main conference track or to the special track on application of Large Language Models focussing on the utilisation of LLMs and language-model approaches for the analysis and exploitation of textual resources in Linguistics, Computer Science, Information Science, the Humanities and the Social Sciences. We welcome the following types of paper submissions: * Long papers (8 pages plus references), describing original research with substantial new results. * Short papers (4 pages plus references), including small focused contributions, work in progress, as well as descriptions of projects, systems and resources. * Abstracts (1 page, non-archival). Abstracts can be presented at the poster session but will not be included in the conference proceedings. We especially invite submission on ongoing projects, student projects, past or ongoing bachelor and master theses, ongoing or recently completed PhD theses, and opinion pieces in this category to foster interaction and discussion in our community. Accepted papers will be presented orally or as posters as determined by the program chairs. However, a preference for a poster or oral presentation can be given. All presentations will take place in person (no online/hybrid format). The main conference language is English. Contributions written in both German or English will be accepted, however, English is recommended. The review process for long and short papers will be double-blind. Submissions of long and short papers must be anonymized accordingly. The submission of non-archival abstracts must not be anonymous (review of abstracts is single-blind). The conference proceedings will be published via the ACL Anthology (archival submissions only). Papers must be presented at the conference by at least one of the authors in order to be published as part of the proceedings. Papers must be formatted in accordance with the ACL style sheet: https://github.com/acl-org/acl-style-files. We strongly encourage authors to use LaTeX in preparing their document. Important Dates Paper Submission Deadline: May 3, 2025 Paper Notification: June 27, 2025 Camera Ready: August 15, 2025 Main Conference: September 10 - 12, 2025 Best whishes Christian Wartena Ulrich Heid Prof. Dr. Christian Wartena Hochschule Hannover Fakultät III - Medien, Information und Design Abt. Information und Kommunikation Lehrgebiet Sprach- und Wissensverarbeitung Expo Plaza 12 30539 Hannover e-mail: christian.wartena(a)hs-hannover.de<mailto:christian.wartena@hs-hannover.de> [DATA-H-Logo_RGB_Unterzeile_klein]

1 0

Call for Participation: 2nd LLMs4Subjects Shared Task
by D'Souza, Jennifer 11 Mar '25

11 Mar '25

The 2nd LLMs4Subjects Shared Task: LLM-based Subject Tagging for the TIB Technical Library's Open-Access Catalog Theme: The Development of Energy- and Compute-Efficient LLM Systems Organized as part of the German Evaluation (GermEval 2025) Shared Task Series 10. - 12. September, 2025 Hildesheim, Germany (co-located with KONVENS 2025 - Conference on Natural Language Processing) 2nd LLMs4Subjects Shared Task: https://sites.google.com/view/llms4subjects-germeval/ KONVENS 2025: https://konvens-2025.hs-hannover.de/about/ Task Overview LLMs4Subjects challenges the research community to develop cutting-edge LLM-based solutions for subject tagging of technical records from Leibniz University's Technical Library (TIBKAT). Participants are tasked with leveraging large language models (LLMs) to tag technical records using the GND taxonomy. The task involves bilingual language modeling, as systems must process technical documents in both German and English. Successful solutions may be integrated into the operational workflows of TIB, the Leibniz Information Centre for Science and Technology. With the rapid advancements in LLMs, the focus is shifting toward making these models more energy- and compute-efficient while maintaining high performance. Recent innovations, such as the DeepSeek series, have demonstrated how techniques like mixture-of-experts (MoE) and model distillation can significantly reduce computational costs without sacrificing effectiveness. The 2nd LLMs4Subjects shared task highlights the importance of efficiency in LLMs, encouraging participants to explore strategies that enhance model performance while optimizing for energy consumption and inference speed. We welcome approaches (but not limited to) that leverage model compression, quantization, efficient fine-tuning, and adaptive computation techniques to push the boundaries of sustainable AI development. Subtasks The 2nd LLMs4Subjects shared task organizes the following two subtasks: Subtask 1 - Multi-Domain Classification of Library Records Subtask 2 - Large-scale Multilabel Subject Indexing of Library Records Important Dates * Release of training data: March 8, 2025 * Release of testing data: May 23, 2025 * Deadline for system submissions: June 2, 2025 * Evaluation end: June 27, 2025 * Paper submission deadline: July 7, 2025 * Notification of acceptance: June 28, 2025 * Camera-ready paper due: August 15, 2025 * Workshop/KONVENS: September 10 - 12, 2025 (TBA)

1 0

2026

2025

2024

2023

2022

Corpora