Two Papers Accepted by ICSC 2025

6 February 2025, by Longquan Jiang

We are happy to announce that two papers from SEMS have been accepted at ICSC 2025 in Los Angeles, CA, USA. 

• Jiang, L., Huang, J., Möller, C. Usbeck, R. (2025): Ontology-Guided, Hybrid Prompt Learning for Generalization in Knowledge Graph Question Answering. ICSC 2025, Los Angeles CA, USA. (pdf)
  Abstract: Most existing Knowledge Graph Question Answering (KGQA) approaches are designed for a specific KG, such as Wikidata, DBpedia or Freebase. Due to the heterogeneity of the underlying graph schema, topology and assertions, most KGQA systems cannot be transferred to unseen Knowledge Graphs (KGs) without resource-intensive training data. We present OntoSCPrompt, a novel Large Language Model (LLM)-based KGQA approach with a two-stage architecture that separates semantic parsing from KG-dependent interactions. OntoSCPrompt first generates a SPARQL query structure (including SPARQL keywords such as SELECT, ASK, WHERE and placeholders for missing tokens) and then fills them with KG-specific information. To enhance the understanding of the underlying KG, we present an ontology-guided, hybrid prompt learning strategy that integrates KG ontology into the learning process of hybrid prompts (e.g., discrete and continuous vectors). We also present several task-specific decoding strategies to ensure the correctness and executability of generated SPARQL queries in both stages. Experimental results demonstrate that OntoSCPrompt performs as well as SOTA approaches without retraining on a number of KGQA datasets such as CWQ, WebQSP and LC-QuAD 1.0 in a resource-efficient manner and can generalize well to unseen domain-specific KGs like DBLP-QuAD and CoyPu KG.
• Taffa, T. and Usbeck, R. (2025): SH-CoDE: Scholarly Hybrid Complex Question Decomposition and Execution. ICSC 2025, Los Angeles CA, USA. (pdf)
  Abstract: Our research addresses the challenge of answering complex scholarly hybrid questions, often demanding multi-faceted reasoning and iterative answer retrieval over scholarly knowledge graphs (KGs) and text. The question complexity is simplified by decomposing it into simple questions and utilizing symbolic representation. However, existing scholarly hybrid Question Answering (QA) models lack question decomposition and symbolic representation. In response, we propose SH-CoDE (Scholarly Hybrid Complex Question Decomposition and Execution). This approach breaks down questions into simple queries and employs symbolic representations, resulting in a natural and interpretable format - HQ (Hybrid Question) representation. SH-CoDE also includes an HQ-Executor, transforming the HQ representation into a tree structure and executing operations within its nodes. During execution, if the executor encounters symbolic representations such as KGQA or TextQA, it retrieves answers from KG and text data sources, respectively. The KGQA module automatically generates and runs SPARQL queries against the KG SPARQL endpoints. Similarly, the TextQA component employs semantic searching and an FLAN-T5-based reader to answer over text. Our model demonstrates competitive results on the test dataset, showcasing its effectiveness in answering complex scholarly questions.