Bohui Zhang

Ontological knowledge underpins large language model (LLM) performance on knowledge-intensive tasks. However, existing evaluations of LLMs’ ontological understanding largely rely on behavioral correctness, leaving the underlying mechanisms unexplained. Existing mechanistic interpretability work has largely focused on instance-level factual knowledge, leaving ontological knowledge under-explored. To bridge this gap, we introduce MIOE, a causal-abstraction benchmark that evaluates interchange interventions for two ontology engineering tasks: entity typing and subsumption prediction. MIOE tests whether learned internal intervention targets can change an ontology-level target answer while preserving class-specific auxiliary information. Across Qwen3-8B and Qwen3-4B, we evaluate SAE-, DBM-, and DAS-based methods, including multi-task and multi-layer variants. Distributed subspace methods give the strongest results and generalize across prompt settings, but the induced changes remain only partially separated from auxiliary class-specific information. Multi-task objectives and contiguous layer-window interventions improve this balance, while class-level analyses reveal substantial variation across ontology classes. These results suggest that ontological predictions are mechanistically accessible in LLMs, although current methods do not yet isolate a cleanly disentangled causal mechanism for ontological knowledge.

Schemas play a vital role in ensuring data quality and supporting usability in the Semantic Web and natural language processing. Traditionally, their creation demands substantial involvement from knowledge engineers and domain experts. Leveraging the impressive capabilities of large language models (LLMs) in tasks like ontology engineering, we explore schema generation using LLMs. To bridge the resource gap, we introduce two datasets: YAGO Schema and Wikidata EntitySchema, along with novel evaluation metrics. The LLM-based pipelines utilize local and global information from knowledge graphs (KGs) to generate schemas in Shape Expressions (ShEx). Experiments demonstrate LLMs’ strong potential in producing high-quality ShEx schemas, paving the way for scalable, automated schema generation for large KGs. Furthermore, our benchmark introduces a new challenge for structured generation, pushing the limits of LLMs on syntactically rich formalisms.

Knowledge graphs (KGs) are crucial in human-centered AI as they provide large labeled machine learning datasets, enhance retrieval-augmented generation, and generate explanations. However, knowledge graph construction has evolved into a complex, semi-automatic process that increasingly relies on black-box deep learning models and heterogeneous data sources to scale. The knowledge graph lifecycle is not transparent, accountability is limited, and there are no accounts of, or methods to determine, how fair a knowledge graph is in downstream applications. KGs are thus at odds with AI regulation, for instance, the EU’s AI Act, and with efforts elsewhere in AI to audit and debias data and algorithms. This paper reports on work towards designing explainable human-in-the-loop knowledge graph construction pipelines. Our work is based on a systematic literature review, in which we study tasks in knowledge graph construction that are often automated, as well as methods to explain how they work and their outcomes, and an interview study with 13 experts from the knowledge engineering community. To analyze the literature, we introduce use cases, related goals for explainable AI (XAI) methods in knowledge graph construction, and the gaps in each use case. To understand the role of XAI models in practice and reveal requirements for improving current methods, we designed interview questions covering broad transparency topics, along with example discussion sessions using examples from the review. From practical knowledge engineering experience, we identify user requirements, propose design blueprints, and outline directions for future research.

Ontology engineering (OE) in large projects poses a number of challenges arising from the heterogeneous backgrounds of the various stakeholders, domain experts, and their complex interactions with ontology designers. This multi-party interaction often creates systematic ambiguities and biases from the elicitation of ontology requirements, which directly affect the design, evaluation and may jeopardise the target reuse. Meanwhile, current OE methodologies strongly rely on manual activities (e.g., interviews, discussion pages). After collecting evidence on the most crucial OE activities, we introduce \textbfOntoChat, a framework for conversational ontology engineering that supports requirement elicitation, analysis, and testing. By interacting with a conversational agent, users can steer the creation of user stories and the extraction of competency questions, while receiving computational support to analyse the overall requirements and test early versions of the resulting ontologies. We evaluate OntoChat by replicating the engineering of the Music Meta Ontology, and collecting preliminary metrics on the effectiveness of each component from users. We release all code at https://github.com/King-s-Knowledge-Graph-Lab/OntoChat.