RAGE-KG1 is an academic venue for outstanding research and bold proposals that integrate retrievalaugmented generation (RAG) with knowledge graphs (KGs). As generative AI continues to mature, research at the intersection of language modelling (LM) and knowledge representation (KR) is driven by the need to produce reliable, verifiable, and context-aware responses grounded in structured, decentralised, and authoritative data sources. By uniting symbolic and subsymbolic approaches, RAG systems can address majour shortcomings of generative AI, strengthening trust in AI and fostering interpretability of AI pipelines. The importance of this research area today is evident in its widespread adoption across both academic research and industrial applications. Over 29% of all Research Track contributions accepted at ISWC 2025 [1] focus on integrating semantic technologies with LLMs, while an impressive 73% of Industry Track contributions2 target this intersection, emphasising its wide and growing practical appeal.
(N. Mihindukulasooriya); https://people.kmi.open.ac.uk/francesco/ (F. Osborne) CEUR
ceur-ws.org neural language models.
In 2025,3 the workshop received 21 submissions, of which 15 were accepted for inclusion in these proceedings. Each contribution received 3–4 reviews during the peer review process. We thank the authors for their high-quality contributions, the programme committee for their diligent reviews, and all participants for their thoughtful engagement in discussions throughout the workshop.
The workshop is honoured to feature a keynote presentation by Roberto Navigli, full professor4 of Natural Language Processing at Sapienza University of Rome, Fellow of AAAI, ACL, EurAI, and ELLIS. Roberto Navigli is the creator of BabelNet,5 the largest multilingual encyclopedic computational dictionary, and co-founder of Babelscape,6 which focuses on multilingual Natural Language Understanding. His pioneering work in knowledge graphs and semantic technologies positions him at the forefront of GraphRAG research, bridging the gap between structured knowledge and large language models. He leads the Minerva LLM family project,7 the first pretrained LLM in Italian, and has received prestigious ERC grants for his groundbreaking research in AI and NLP.
BabelNet, NounAtlas, Concept-pedia, and Other Marvels: Exploring Semantics in the Age of LLMs
Large Language Models (LLMs) have redefined the distributional paradigm in semantics, demonstrating that large-scale statistical learning can yield emergent representations of meaning. Yet, while these models exhibit impressive linguistic fluency and versatility, their internal representations of meaning remain largely opaque, data-driven, and detached from explicit conceptual structure. This talk revisits the problem of meaning representation from a complementary, knowledge-based perspective, presenting an integrated view of several large-scale semantic resources — including BabelNet, NounAtlas, and Concept-pedia — that aim to provide interpretable, multilingual, and multimodal conceptually-grounded frameworks for modeling lexical and conceptual knowledge.
We will also discuss the potential of explicit semantics to interface with LLMs for enhanced interpretability and semantic alignment. In doing so, the talk argues for a renewed synthesis between symbolic and subsymbolic approaches to meaning, illustrating how curated, multilingual knowledge graphs and data-driven models can jointly contribute to a more comprehensive and transparent account of semantics in the era of large-scale neural language modelling.
• Sergio José Rodríguez Méndez, Australian
National University • Tek Raj Chhetri, Massachusetts Institute of
Technology • Hamed Babaei Giglou, TIB • Axel Polleres, WU Vienna • Margherita Martorana, VU Amsterdam • Gaetano Rossiello, IBM Research • Reham Alharbi, University of Liverpool • Laura Menotti, University of Padova 3See: https://2025.rage-kg.org/ 4See: http://www.diag.uniroma1.it/navigli/ 5See: https://babelnet.org 6See: https://babelscape.com 7See: https://minerva-ai.org • Anouk Oudshoorn, TU Vienna • Edgard Marx, Leipzig University of Applied
Sciences • Kabul Kurniawan, Universitas Gadjah
Mada • Sven Groppe, University of Lübeck • George Hannah, University of Liverpool • Vincenzo Nucci, University of Camerino • Arianna Fedeli, Gran Sasso Science Institute • Fatima Zahra Amara, University of Bari
Aldo Moro • Azanzi Jiomekong, University of Yaounde • Sola Shirai, IBM Research • Sameer Sadruddin, TIB • Johan Cederbladh, Mälardalen University
This workshop was funded in whole or in part by the Austrian Science Fund (FWF) 10.55776/COE12, and by the COST Action CA23147 GOBLIN - Global Network on Large-Scale, Cross-domain and Multilingual Open Knowledge Graphs, supported by COST (European Cooperation in Science and Technology, https://www.cost.eu).