@article{60990, abstract = {{Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of natural language processing tasks. However, their effectiveness in low-resource languages remains underexplored, particularly in complex tasks such as end-to-end Entity Linking (EL), which requires both mention detection and disambiguation against a knowledge base (KB). In earlier work, we introduced IndEL — the first end-to-end EL benchmark dataset for the Indonesian language — covering both a general domain (news) and a specific domain (religious text from the Indonesian translation of the Quran), and evaluated four traditional end-to-end EL systems on this dataset. In this study, we propose ELEVATE-ID, a comprehensive evaluation framework for assessing LLM performance on end-to-end EL in Indonesian. The framework evaluates LLMs under both zero-shot and fine-tuned conditions, using multilingual and Indonesian monolingual models, with Wikidata as the target KB. Our experiments include performance benchmarking, generalization analysis across domains, and systematic error analysis. Results show that GPT-4 and GPT-3.5 achieve the highest accuracy in zero-shot and fine-tuned settings, respectively. However, even fine-tuned GPT-3.5 underperforms compared to DBpedia Spotlight — the weakest of the traditional model baselines — in the general domain. Interestingly, GPT-3.5 outperforms Babelfy in the specific domain. Generalization analysis indicates that fine-tuned GPT-3.5 adapts more effectively to cross-domain and mixed-domain scenarios. Error analysis uncovers persistent challenges that hinder LLM performance: difficulties with non-complete mentions, acronym disambiguation, and full-name recognition in formal contexts. These issues point to limitations in mention boundary detection and contextual grounding. Indonesian-pretrained LLMs, Komodo and Merak, reveal core weaknesses: template leakage and entity hallucination, respectively—underscoring architectural and training limitations in low-resource end-to-end EL.11Code and dataset are available at https://github.com/dice-group/ELEVATE-ID.}}, author = {{Gusmita, Ria Hari and Firmansyah, Asep Fajar and Zahera, Hamada Mohamed Abdelsamee and Ngonga Ngomo, Axel-Cyrille}}, issn = {{0169-023X}}, journal = {{Data & Knowledge Engineering}}, keywords = {{LLMs, Evaluation, End-to-end EL, Indonesian}}, pages = {{102504}}, title = {{{ELEVATE-ID: Extending Large Language Models for End-to-End Entity Linking Evaluation in Indonesian}}}, doi = {{https://doi.org/10.1016/j.datak.2025.102504}}, volume = {{161}}, year = {{2026}}, } @inproceedings{57324, abstract = {{Generating SPARQL queries is crucial for extracting relevant information from diverse knowledge graphs. However, the structural and semantic differences among these graphs necessitate training or fine-tuning a tailored model for each one. In this paper, we propose UniQ-Gen, a unified query generation approach to generate SPARQL queries across various knowledge graphs. UniQ-Gen integrates entity recognition, disambiguation, and linking through a BERT-NER model and employs cross-encoder ranking to align questions with the Freebase ontology. We conducted several experiments on different benchmark datasets such as LC-QuAD 2.0, GrailQA, and QALD-10. The evaluation results demonstrate that our approach achieves performance equivalent to or better than models fine-tuned for individual knowledge graphs. This finding suggests that fine-tuning a unified model on a heterogeneous dataset of SPARQL queries across different knowledge graphs eliminates the need for separate models for each graph, thereby reducing resource requirements.}}, author = {{Vollmers, Daniel and Srivastava, Nikit and Zahera, Hamada Mohamed Abdelsamee and Moussallem, Diego and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Knowledge Engineering and Knowledge Management}}, editor = {{Alam, Mehwish and Rospocher, Marco and van Erp, Marieke and Hollink, Laura and Gesese, Genet Asefa}}, isbn = {{978-3-031-77792-9}}, pages = {{174–189}}, publisher = {{Springer Nature Switzerland}}, title = {{{UniQ-Gen: Unified Query Generation Across Multiple Knowledge Graphs}}}, doi = {{https://doi.org/10.1007/978-3-031-77792-9_11}}, year = {{2025}}, } @inproceedings{59054, author = {{Firmansyah, Asep Fajar and Zahera, Hamada Mohamed Abdelsamee and Sherif, Mohamed and Moussallem, Diego and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{ESWC2025}}, isbn = {{978-3-031-94575-5}}, keywords = {{firmansyah mousallem ngonga sherif zahera}}, pages = {{133----151}}, publisher = {{pringer Nature Switzerland}}, title = {{{ANTS: Abstractive Entity Summarization in Knowledge Graphs}}}, doi = {{10.1007/978-3-031-94575-5_8}}, year = {{2025}}, } @article{61134, author = {{Manzoor, Ali and Speck, René and Zahera, Hamada Mohamed Abdelsamee and Saleem, Muhammad and Moussallem, Diego and Ngonga Ngomo, Axel-Cyrille}}, issn = {{2169-3536}}, journal = {{IEEE Access}}, pages = {{1--1}}, publisher = {{Institute of Electrical and Electronics Engineers (IEEE)}}, title = {{{Multilingual Relation Extraction - A Survey}}}, doi = {{10.1109/access.2025.3604258}}, year = {{2025}}, } @inproceedings{61041, abstract = {{Large Language Models (LLMs) are increasingly deployed in real-world applications that require access to up-to-date knowledge. However, retraining LLMs is computationally expensive. Therefore, knowledge editing techniques are crucial for maintaining current information and correcting erroneous assertions within pre-trained models. Current benchmarks for knowledge editing primarily focus on recalling edited facts, often neglecting their logical consequences. To address this limitation, we introduce a new benchmark designed to evaluate how knowledge editing methods handle the logical consequences of a single fact edit. Our benchmark extracts relevant logical rules from a knowledge graph for a given edit. Then, it generates multi-hop questions based on these rules to assess the impact on logical consequences. Our findings indicate that while existing knowledge editing approaches can accurately insert direct assertions into LLMs, they frequently fail to inject entailed knowledge. Specifically, experiments with popular methods like ROME and FT reveal a substantial performance gap, up to 24%, between evaluations on directly edited knowledge and on entailed knowledge. This highlights the critical need for semantics-aware evaluation frameworks in knowledge editing.}}, author = {{Moteu Ngoli, Tatiana and Kouagou, N'Dah Jean and Zahera, Hamada Mohamed Abdelsamee and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Proceedings of the 24th International Semantic Web Conference (ISWC 2025)}}, isbn = {{978-3-032-09530-5}}, keywords = {{dice sailproject moteu kouagou zahera ngonga}}, location = {{Nara, Japan}}, pages = {{pp 41--56}}, publisher = {{Springer, Cham}}, title = {{{Benchmarking Knowledge Editing using Logical Rules}}}, doi = {{https://doi.org/10.1007/978-3-032-09530-5_3}}, year = {{2025}}, } @inproceedings{61753, abstract = {{This paper presents LOLA, a massively multilingual large language model trained on more than 160 languages using a sparse Mixture-of-Experts Transformer architecture. Our architectural and implementation choices address the challenge of harnessing linguistic diversity while maintaining efficiency and avoiding the common pitfalls of multilinguality. Our analysis of the evaluation results shows competitive performance in natural language generation and understanding tasks. Additionally, we demonstrate how the learned expert-routing mechanism exploits implicit phylogenetic linguistic patterns to potentially alleviate the curse of multilinguality. We provide an in-depth look at the training process, an analysis of the datasets, and a balanced exploration of the model{’}s strengths and limitations. As an open-source model, LOLA promotes reproducibility and serves as a robust foundation for future research. Our findings enable the development of compute-efficient multilingual models with strong, scalable performance across languages.}}, author = {{Srivastava, Nikit and Kuchelev, Denis and Moteu Ngoli, Tatiana and Shetty, Kshitij and Röder, Michael and Zahera, Hamada Mohamed Abdelsamee and Moussallem, Diego and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Proceedings of the 31st International Conference on Computational Linguistics}}, editor = {{Rambow, Owen and Wanner, Leo and Apidianaki, Marianna and Al-Khalifa, Hend and Eugenio, Barbara Di and Schockaert, Steven}}, pages = {{6420–6446}}, publisher = {{Association for Computational Linguistics}}, title = {{{LOLA – An Open-Source Massively Multilingual Large Language Model}}}, year = {{2025}}, } @inproceedings{54449, author = {{KOUAGOU, N'Dah Jean and Demir, Caglar and Zahera, Hamada Mohamed Abdelsamee and Wilke, Adrian and Heindorf, Stefan and Li, Jiayi and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Companion Proceedings of the ACM on Web Conference 2024}}, location = {{Singapore}}, publisher = {{ACM}}, title = {{{Universal Knowledge Graph Embeddings}}}, doi = {{10.1145/3589335.3651978}}, year = {{2024}}, } @unpublished{57277, author = {{Srivastava, Nikit and Ma, Mengshi and Vollmers, Daniel and Zahera, Hamada Mohamed Abdelsamee and Moussallem, Diego and Ngonga Ngomo, Axel-Cyrille}}, title = {{{MST5 -- Multilingual Question Answering over Knowledge Graphs}}}, year = {{2024}}, } @inproceedings{55094, author = {{Zahera, Hamada Mohamed Abdelsamee and Manzoor, Ali and Sherif, Mohamed and Moussallem, Diego and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{SEMANTiCS}}, keywords = {{TRR318 climatebowl colide dice enexa kiam manzoor moussallem ngonga sailproject sherif simba zahera}}, title = {{{Generating SPARQL from Natural Language Using Chain-of-Thoughts Prompting}}}, year = {{2024}}, } @inproceedings{54608, author = {{Zahera, Hamada Mohamed Abdelsamee and Vitiugin, Fedor and Sherif, Mohamed and Castillo, Carlos and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{SEMANTiCS}}, keywords = {{dice kiam ngonga porque sherif zahera}}, title = {{{Using Pre-trained Language Models for Abstractive DBPEDIA Summarization: A Comparative Study}}}, year = {{2023}}, } @inproceedings{46518, abstract = {{Purpose: This study addresses the limitations of current short abstracts of DBpedia entities, which often lack a comprehensive overview due to their creating method (i.e., selecting the first two-three sentences from the full DBpedia abstracts). Methodology: We leverage pre-trained language models to generate abstractive summaries of DBpedia abstracts in six languages (English, French, German, Italian, Spanish, and Dutch). We performed several experiments to assess the quality of generated summaries by language models. In particular, we evaluated the generated summaries using human judgments and automated metrics (Self-ROUGE and BERTScore). Additionally, we studied the correlation between human judgments and automated metrics in evaluating the generated summaries under different aspects: informativeness, coherence, conciseness, and fluency. Findings: Pre-trained language models generate summaries more concise and informative than existing short abstracts. Specifically, BART-based models effectively overcome the limitations of DBpedia short abstracts, especially for longer ones. Moreover, we show that BERTScore and ROUGE-1 are reliable metrics for assessing the informativeness and coherence of the generated summaries with respect to the full DBpedia abstracts. We also find a negative correlation between conciseness and human ratings. Furthermore, fluency evaluation remains challenging without human judgment. Value: This study has significant implications for various applications in machine learning and natural language processing that rely on DBpedia resources. By providing succinct and comprehensive summaries, our approach enhances the quality of DBpedia abstracts and contributes to the semantic web community}}, author = {{Zahera, Hamada Mohamed Abdelsamee and Vitiugin, Fedor and Sherif, Mohamed and Castillo, Carlos and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{SEMANTiCS}}, keywords = {{dice enexa kiam ngonga porque sherif zahera}}, location = {{Leipzig, Germany}}, title = {{{Using Pre-trained Language Models for Abstractive DBpedia Summarization: A Comparative Study}}}, year = {{2023}}, } @inbook{33738, author = {{Zahera, Hamada Mohamed Abdelsamee and Heindorf, Stefan and Balke, Stefan and Haupt, Jonas and Voigt, Martin and Walter, Carolin and Witter, Fabian and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{The Semantic Web: ESWC 2022 Satellite Events}}, isbn = {{9783031116087}}, issn = {{0302-9743}}, publisher = {{Springer International Publishing}}, title = {{{Tab2Onto: Unsupervised Semantification with Knowledge Graph Embeddings}}}, doi = {{10.1007/978-3-031-11609-4_9}}, year = {{2022}}, } @inproceedings{46538, author = {{Zahera, Hamada Mohamed Abdelsamee and Vollmers, Daniel and Sherif, Mohamed and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{ISWC 2022}}, isbn = {{978-3-031-19432-0}}, keywords = {{colide dice eml4u ngonga raki sherif speaker vollmers zahera}}, publisher = {{Springer, Cham}}, title = {{{MultPAX: Keyphrase Extraction using Language Models and Knowledge Graphs}}}, doi = {{10.1007/978-3-031-19433-7_18}}, year = {{2022}}, } @inproceedings{29291, author = {{Zahera, Hamada Mohamed Abdelsamee and Heindorf, Stefan and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{Proceedings of the 11th on Knowledge Capture Conference}}, publisher = {{ACM}}, title = {{{ASSET: A Semi-supervised Approach for Entity Typing in Knowledge Graphs}}}, doi = {{10.1145/3460210.3493563}}, year = {{2021}}, } @inproceedings{29043, abstract = {{Social media plays a significant role in disaster management by providing valuable data about affected people, donations and help requests. Recent studies highlight the need to filter information on social media into fine-grained content labels. However, identifying useful information from massive amounts of social media posts during a crisis is a challenging task. In this paper, we propose I-AID, a multimodel approach to automatically categorize tweets into multi-label information types and filter critical information from the enormous volume of social media data. I-AID incorporates three main components: i) a BERT- based encoder to capture the semantics of a tweet and represent as a low-dimensional vector, ii) a graph attention network (GAT) to apprehend correlations between tweets’ words/entities and the corresponding information types, and iii) a Relation Network as a learnable distance metric to compute the similarity between tweets and their corresponding information types in a supervised way. We conducted several experiments on two real publicly-available datasets. Our results indicate that I-AID outperforms state-of- the-art approaches in terms of weighted average F1 score by +6% and +4% on the TREC-IS dataset and COVID-19 Tweets, respectively.}}, author = {{Zahera, Hamada Mohamed Abdelsamee and Jalota, Rricha and Sherif, Mohamed and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{IEEE Open Access}}, keywords = {{sys:relevantFor:infai sys:relevantFor:DAIKIRI ngonga zahera sherif daikiriproject dice simba}}, title = {{{I-AID: Identifying Actionable Information from Disaster-related Tweets}}}, year = {{2021}}, } @inproceedings{29044, author = {{Chakraborty, Jaydeep and Sherif, Mohamed and Zahera, Hamada Mohamed Abdelsamee and Bansal, Srividya}}, booktitle = {{Proceedings of the IEEE International Conference on Machine Learning and Applications}}, keywords = {{dice sherif hamada}}, title = {{{OntoConnect: Domain-Agnostic Ontology Alignment using Graph Embedding with Negative Sampling}}}, year = {{2021}}, } @inproceedings{29040, author = {{Zahera, Hamada Mohamed Abdelsamee and Sherif, Mohamed}}, booktitle = {{Proceedings of Mining the Web of HTML-embedded Product Data Workshop (MWPD2020)}}, keywords = {{2020 dice zahera sherif knowgraphs sys:relevantFor:limboproject limboproject sys:relevantFor:infai sys:relevantFor:bis limes limbo opal}}, title = {{{ProBERT: Product Data Classification with Fine-tuning BERT Model}}}, year = {{2020}}, } @inproceedings{29037, abstract = {{Existing technologies employ different machine learning approaches to predict disasters from historical environmental data. However, for short-term disasters (e.g., earthquakes), historical data alone has a limited prediction capability. In this work, we consider social media as a supplementary source of knowledge in addition to historical environmental data. Further, we build a joint model that learns from disaster-related tweets and environmental data to improve prediction. We propose the combination of semantically-enriched word embedding to represent entities in tweets with their semantics representations computed with the traditional word2vec. Our experiments show that our proposed approach outperforms the accuracy of state-of-the-art models in disaster prediction.}}, author = {{Zahera, Hamada Mohamed Abdelsamee and Sherif, Mohamed and Ngonga Ngomo, Axel-Cyrille}}, booktitle = {{K-CAP 2019: Knowledge Capture Conference}}, keywords = {{sys:relevantFor:infai sys:relevantFor:bis sys:relevantFor:simba ngonga simba zahera sherif solide limboproject opal group\_aksw dice}}, pages = {{4}}, title = {{{Jointly Learning from Social Media and Environmental Data for Typhoon Intensity Prediction}}}, year = {{2019}}, } @inproceedings{29003, abstract = {{In this paper, we describe our approach to classify disaster-related tweets into multilabel information types (ie, labels). We aim to filter first relevant tweets during disasters. Then, we assign tweets relevant information types. Information types can be SearchAndRescue, MovePeople and Volunteer. We employ a fine-tuned BERT model with 10 BERT layers. Further, we submitted our approach to the TREC-IS 2019 challenge, the evaluation results showed that our approach outperforms the F1-score of median score in identifying actionable information.}}, author = {{Zahera, Hamada Mohamed Abdelsamee and A. Elgendy, Ibrahim and Jalota, Rricha and Sherif, Mohamed}}, booktitle = {{Proceedings of the Twenty-Eighth Text REtrieval Conference, {TREC} 2019, Gaithersburg, Maryland, USA, November 13-15, 2019}}, keywords = {{zahera elgendy jalota sherif dice}}, title = {{{Fine-tuned BERT Model for Multi-Label Tweets Classification}}}, year = {{2019}}, }