--------------------------------------------------------------------------------- Workshop on Generative AI and Knowledge Graphs (GenAIK), 19 January 2025, Abu Dhabi, UAE Web: https://genetasefa.github.io/GenAIK2025/ X: @GenAIK25 LinkedIn: https://www.linkedin.com/groups/9868047 Mastodon: https://sigmoid.social/@GenAIK --------------------------------------------------------------------------------- In conjunction with COLING 2025, January 19-24 --------------------------------------------------------------------------------- Workshop Overview --------------------------------------------------------------------------------- Generative Artificial Intelligence (GenAI) is a branch of artificial intelligence capable of creating seemingly new, meaningful content, including text, images, and audio. It utilizes deep learning models, such as Large Language Models (LLMs), to recognize and replicate data patterns, enabling the generation of human-like content. Notable families of LLMs include GPT (GPT-3.5, GPT-3.5 Turbo, and GPT-4), LLaMA (LLaMA and LLaMA-2), and Mistral (Mistral and Mixtral). GPT, which stands for Generative Pretrained Transformer, is especially popular for text generation and is widely used in applications like ChatGPT. GenAI has taken the world by storm and revolutionized various industries, including healthcare, finance, and entertainment. However, GenAI models have several limitations, including biases from training data, generating factually incorrect information, and difficulty in understanding complex content. Additionally, their performance can vary based on domain specificity. In recent times, Knowledge Graphs (KGs) have attracted considerable attention for their ability to represent structured and interconnected information, and adopted by many companies in various domains. KGs represent knowledge by depicting relationships between entities, known as facts, usually based on formal ontological models. Consequently, they enable accuracy, decisiveness, interpretability, domain-specific knowledge, and evolving knowledge in various AI applications. The intersection between GenAI and KG has ignited significant interest and innovation in Natural Language Processing (NLP). For instance, by integrating LLMs with KGs during pre-training and inference, external knowledge can be incorporated for enhancing the model’s capabilities and improving interpretability. When integrated, they offer a robust approach to problem solving in diverse areas such as information enrichment, representation learning, conversational AI, cross-domain AI transfer, bias, content generation, and semantic understanding. This workshop aims at reinforcing the relationships between Deep Learning, Knowledge Graphs, and NLP communities and foster interdisciplinary research in the area of GenAI. --------------------------------------------------------------------------------- Topics of Interest --------------------------------------------------------------------------------- * Enhancing KG construction and completion with GenAI * Multimodal KG generation * Text-to-KG using LLMs * Multilingual KGs * GenAI for KG embeddings * GenAI for Temporal KGs * Dialogue systems enhanced by KG and GenAI * Cross-domain knowledge transfer with GenAI * Bias mitigation using KGs in GenAI * Explainability with KGs and GenAI * Natural language querying of KGs via GenAI * NLP tasks using KGs and GenAI * Prompt Engineering using KGs * GenAI for Ontology learning and schema induction in KGs * Hybrid QA systems combining KGs and GenAI * Recommendation systems and KGs with GenAI * Creating benchmark datasets relevant for tasks combining KGs and GenAI * Real-world applications on scholarly data, biomedical domain, etc. * Knowledge Graph Alignment * Applying to real-world scenarios ------------------------------------------------------------------------------------ Important Dates ------------------------------------------------------------------------------------ - Submission deadline: 5 November 2024 - Notification of Acceptance: 5 December 2024 - Camera-ready paper due: 13 December 2024 - COLING2025 Workshop day: 19 January 2025 ------------------------------------------------------------------------------------ Submissions ------------------------------------------------------------------------------------ Full research papers (6-8 pages) Short research papers (4-6 pages) Position papers (2 pages) These page limits only apply to the main body of the paper. At the end of the paper (after the conclusions but before the references) papers need to include a mandatory section discussing the limitations of the work and, optionally, a section discussing ethical considerations. Papers can include unlimited pages of references and an unlimited appendix. Papers must follow the two-column format of *ACL conferences, using the official templates (https://www.overleaf.com/latex/templates/association-for-computational-lingu... ). The templates are available for download as style files and formatting guidelines. Submissions that do not adhere to the specified styles, including paper size, font size restrictions, and margin width, will be desk-rejected. Submissions are open to all and must be anonymous, adhering to COLING 2025's double-blind submission and reproducibility guidelines. All accepted papers (after double-blind review of at least 3 experts) will appear in the workshop proceedings that will be published in ACL Anthology. At least one of the authors of the accepted papers must register for the workshop to be included into the workshop proceedings. The workshop will be a 100% in-person 1-day event at COLING 2025. Submissions must be made using the START portal: https://softconf.com/coling2025/GenAIK25/ --------------------------------------------------------------------------------- Sponsors --------------------------------------------------------------------------------- NFDI4DataScience (NFDI4DS - https://www.nfdi4datascience.de/ ) is a national research data infrastructure for Data Science and AI project. The overarching objective of the project is the development, establishment, and sustainment of a national research data infrastructure (NFDI) for the Data Science and Artificial Intelligence community in Germany. The vision of NFDI4DS is to support all steps of the complex and interdisciplinary research data lifecycle, including collecting/creating, processing, analyzing, publishing, archiving, and reusing resources in Data Science and Artificial Intelligence. NFDI4ds is offering a total of €2000 in travel grants (€1000 each) to two selected students who will attend and present their work at GenAIK 2025! To be considered, submit your paper to the workshop, and if your paper is accepted, you’ll be eligible for a chance to receive one of the two grants. --------------------------------------------------------------------------------- Organization --------------------------------------------------------------------------------- - Genet Asefa Gesese, FIZ Karlsruhe, KIT, Germany - Harald Sack, FIZ Karlsruhe, KIT, Germany - Heiko Paulheim, University of Mannheim, Germany - Albert Meroño-Peñuela, King’s College London, UK - Lihu Chen, Imperial College London, UK If you have published in ACL conferences previously, and are interested to be part of the program committee of GenAIK2025, please fill in this form: https://forms.gle/t56dP6McD1VJmTfT9 Dr.-Ing. Genet Asefa Gesese Head of Machine Learning Department Information Service Engineering Phone. +49 7247 808 186 Fax. +49 7247 808 78186 FIZ Karlsruhe – Leibniz Institute for Information Infrastructure Hermann-von-Helmholtz-Platz 1 76344 Eggenstein-Leopoldshafen www.fiz-karlsruhe.de ------------------------------------------------------------------------------ FIZ Karlsruhe - Leibniz-Institut für Informationsinfrastruktur GmbH. Sitz der Gesellschaft: Eggenstein-Leopoldshafen, Amtsgericht Mannheim HRB 101892. Geschäftsführer: Prof. Dr. Wolfram Horstmann. Vorsitzende des Aufsichtsrats: MinR’in Marion Steinberger. FIZ Karlsruhe ist zertifiziert mit dem Siegel "audit berufundfamilie".