LLMs encode world knowledge through pre-training on massive datasets, making them the backbone of knowledge extraction tasks. Their reliability degrades on long-tail knowledge: low-popularity knowledge that occurs infrequently in pre-training data. Popularity is not a neutral property: pre-training datasets are predominantly web-crawled and, as such, are generalist, English-centric, and mostly produced over the past 30 years by Western, High-income, Educated, Liberal, Male-dominated (WHELM, Daryani et al 2025) communities, raising the risk of models underperforming on specialised domains, non-English languages and non-contemporary times sources, and on knowledge belonging to marginalised social groups. Retrieval-Augmented Generation has been proposed as a mitigation, but corpora used for retrieval may still be biased. Knowledge Graphs (KGs) provide a more transparent and deterministic alternative, yet open-domain KGs such as Wikidata exhibit coverage gaps along the same dimensions. The X-TAIL workshop aims to advance research on extracting, exploiting, and ultimately preserving long-tail knowledge. It welcomes contributions on: methods to extract knowledge from domain-specific, multilingual, historical, and low-resource language sources, including approaches combining LLMs and KGs; studies on the head/tail knowledge distinction; investigations on how popularity distributes across the specificity, linguistic, temporal, and cultural dimensions of knowledge, and its effect on system performance; characterisation of gaps in knowledge bases and mitigation strategies.
Workshop website: https://www.xtail-workshop.org/
The way legal knowledge is modelled, analysed, and applied is rapidly transforming. This shift is primarily driven by the growing availability of legal data and recent advances in Artificial Intelligence (AI), Knowledge Representation and Reasoning (KRR), Natural Language Processing (NLP), Large Language Models (LLMs), and Process-Oriented Data Science.
To explore these changes, the KM4LAW workshop provides an interdisciplinary forum for researchers and practitioners. We welcome contributions addressing the extraction, representation, interpretation, and use of legal knowledge from structured and unstructured sources. Relevant topics include legal ontologies, knowledge graphs, information extraction, argumentation, explainable AI for law, multilingual legal NLP (including cross-lingual transfer, semantic alignment, and legal translation), and the application of LLMs to tasks like summarisation, classification, and decision support.
In addition, KM4LAW promotes research on legal processes and regulatory compliance through process mining techniques, such as process discovery, conformance checking, and predictive process monitoring. The growing adoption of digital procedures in courts and public administrations creates new opportunities to combine legal knowledge modelling with process-aware methods.
By connecting communities across AI & Law, KRR, NLP, and Business Process Management, KM4LAW fosters collaboration and innovative approaches for the next generation of legal information systems.
Workshop website: https://km4law.di.unito.it/
Organizer contact email:
Rachele Mignone (rachele.mignone@unito.it)
Introducing the 3rd Playing with Meanings workshop (PwM3) which will take place suring the 25th International Conference on Knowledge Engineering and Knowledge Management (EKAW 2026), September 29th - October 1st, 2026, at University of Torino, Italy. The Playing with Meanings 3rd edition (PwM3) continues to develop ontology-driven knowledge co-production methodologies, with new #ontologygames and group modelling activities that address three distinct topics:
i) Knowledge Engineering and Justice Assessment
ii) Bridging Intangible Cultural Heritage and Social-Ecological Systems
iii) Modelling Discourses on Urban Systems and Place-making
Together we will collaboratively generate and examine scenarios, practice group modelling with ontologies, and discuss pluralism and justice, on- tological analysis and shared understandings. A variety of card games, role-play exercises, group modelling activities will be enacted; each inte- grates, teaches, and produces through the use of dedicated domain on- tologies as well as the foundational ontologies UFO, BFO and DOLCE. PwM3 at EKAW2026 will be a full-day workshop, with three dedicated sessions, practicing participatory ontology engineering together.
Workshop website: https://humanfactorsinsemantics.net/PwM3.html
Organizers contact emails:
greta.adamo@unibz.it
max.willis@unibz.it
emiliomaria.sanfilippo@cnr.it
Transformer-based Large Language Models (Tb-LLMs) demonstrate extraordinary capabilities and, thus, change the approach of conducting research even in Knowledge Representation. Building on existing research that neural networks (NNs) on discrete symbols operate as holistic reasoners over “classical” symbolic approaches, this tutorial aims to present the dominant paradigm of Tb-LLMs according to the idea that these models can be understood with neuro-symbolic, ontology-grounded approaches and controlled through mechanistic interpretable or explainable KG reasoning methods. Offering a profound understanding of the inner workings of these models, the tutorial will help practitioners build novel, transparent architectures to model symbolic knowledge in NN.
URL: https://humancentricart.github.io/mechanistic-interpretability-by-design/ekaw/index.html
Legal automation requires not only the ability to represent legal texts in structured formats but also to capture their normative meaning in a way that supports automated reasoning. This tutorial focuses on two key standards Akoma Ntoso for the semantic representation of legal documents and LegalRuleML for modelling legal rules and norms—to enable end-to-end legal automation.
Participants will learn how to transform unstructured legal texts into richly annotated, machine-readable documents using Akoma Ntoso, and how to express obligations, permissions, prohibitions, and exceptions using LegalRuleML. The tutorial will demonstrate how these knowledge based technologies can be combined to build automated compliance systems, support legal reasoning, and deliver explainable outcomes. Real-world use cases from regulatory compliance and policy-driven systems will illustrate how these standards bridge the gap between legal drafting and executable models."