The many faces of GESIS
Vita
M. Taimoor Khan is a senior researcher in the Knowledge Technologies for Social Sciences department at GESIS – Leibniz Institute for the Social Sciences. With a background in computer science, his research focuses on applied machine learning, natural language processing and dynamically adapting models, with computational reproducibility as recent addition. He has also led the development of big data infrastructure and NLP tools for low resourced languages. Khan has contributed to EU project and has published in top-tier computer science venues. At GESIS, he is part of the Methods Hub project aiming to lower the technical barriers in reusing reproducible large computational models.Service
As a technical lead for the Methods Hub project, my responsibilities are:- Supervising Methods Hub portal development and incremental enhancement, as a niche platform for reproducible computational models on social science use cases
- Defining standards and developing resources to assist researchers share their model on the portal
- Collecting, reviewing and developing sample content for the portal
Research
M. Taimoor Khan’s research focuses on advancing technologies for efficiently processing, exploring, and analyzing textual data. His work lies at the intersection of Natural Language Processing (NLP) and Machine Learning (ML), with an emphasis on applied research. He develops innovative models that enhance task performance in controlled environments, while also creating tools that tackle the complexities and inconsistencies of real-world scenarios. Notably, his contributions include adapting large language models for social science applications, such as rumor detection, text detoxification, and sexism detection, in both bilingual and multilingual settings. Additionally, he has explored knowledge transfer across a sequence of tasks within multi-agent systems and addressed the challenges of continual learning, particularly through the expansion of neural networks and mitigating catastrophic forgetting. His work also extends to novelty detection in out-of-distribution cases within the open-world paradigm.Publications
Journal article
Wajid, Usman, Muhammad Hamza, M. Taimoor Khan, and Nouman Azam. 2024. "A Three-way Decision Approach for Dynamically Expandable Networks." International Journal of Approximate Reasoning 166 (March 2024): 109105. doi: https://doi.org/10.1016/j.ijar.2023.109105.
Khan, M. Taimoor, Nouman Azam, Shehzad Khalid, and Furqan Aziz. 2022. "Hierarchical lifelong topic modeling using rules extracted from network communities." PLoS one: Public Library of Science 3 (17): 1-22. doi: https://doi.org/10.1371/journal.pone.0264481.
Chapter in an edited book
Alia, Alaaeddin, and M. Taimoor Khan. 2025 (Forthcoming). "LLM Based Bilingual Rumor Verification Using Evidence From Authorities." In Reducing Online Misinformation through Credible Information Retrieval 2025,
Khan, M. Taimoor, Nasarullah Jan Wazir, Imran Khan, Muhammad Khan Afridi, and Omar Usman Khan. 2025 (Forthcoming). "Out-of-Vocabulary Pashto Spell Checker using Morphological Operations."
Gangopadhyay, Susmita, M. Taimoor Khan, and Hajira Jabeen. 2024. "Linguistic_Hygenist at PAN 2024 TextDetox: HybridDetox - A Combination of Supervised and Unsupervised Methods for Effective Multilingual Text Detoxification." In Working Notes of the Conference and Labs of the Evaluation Forum (CLEF 2024) Grenoble, France, 9-12 September, 2024, edited by Guglielmo Faggioli, Nicola Ferro, Petra Galuščáková, and Alba García Seco de Herrera, CEUR workshop proceedings 3740, 2576-2584. Aachen: RWTH. https://ceur-ws.org/Vol-3740/paper-236.pdf.
Saadi, Khouloud, and M. Taimoor Khan. 2022. "Effective Prevention of Semantic Drift in Continual Deep Learning." In Intelligent Data Engineering and Automated Learning – IDEAL 2022: 23rd International Conference, IDEAL 2022, Manchester, UK, November 24–26, 2022, Proceedings, edited by Hujun Yin, David Camacho, and Peter Tino, Lecture Notes in Computer Science 13756, 456-464. Cham: Springer. doi: https://doi.org/10.1007/978-3-031-21753-1_44. https://link.springer.com/chapter/10.1007/978-3-031-21753-1_44.
Working and discussion paper
Momeni, Fakhri, Eleni Adamidi, Sven Ulpts, Hajira Jabeen, M. Taimoor Khan, Thomas Klebel, Eva Kormann, Barbara Leitner, Allyson Lister, Elli Papadopoulou, Haris Papageorgiou, Tony Ross-Hellauer, Susanna-Assunta Sansone, Jesper W. Schneider, Joeri K. Tijdink, and Thanasis Vergoulis. 2024. TIER2 D4.2 - Pilot Implementation and Assessment Plans. OSF. doi: https://doi.org/10.17605/OSF.IO/GCT56. https://osf.io/e5wmu.
Presentation at a conference
Khan, M. Taimoor, Arnim Bleier, Chung-hong Chan, Po-Chun Chang, Raniere Gaia Costa da Silva, Danilo Dessi, Stefan Dietze, Gabriella Lapesa, Brigitte Mathiak, David Schoch, Claudia Wagner, and Hajira Jabeen. 2024. "Methods Hub: A Platform for Sharing Computational Social Science Methods." 4th Workshop on Computational Linguistics for the Political and Social Sciences (CPSS), 2024-09-13.
Costa da Silva, Raniere Gaia, Arnim Bleier, Chung-hong Chan, Po-Chun Chang, Stefan Dietze, M. Taimoor Khan, Gabriella Lapesa, Brigitte Mathiak, David Schoch, and Claudia Wagner. 2024. "Methods Hub Reproducible Computation Methods for Social Science Research." BioNT Community Event & CarpentryConnect - Heidelberg 2024, 2024-11-12.
Khan, M. Taimoor, Danilo Dessi, Fakhri Momeni, and Hajira Jabeen. 2024. "Reproducibility of AI/ML methods in computational social science." META-REP 2024 – the Conference on Meta-Science and Replicability in the social, behavioral, and cognitive sciences, 2024-10-28.
Bleier, Arnim, Chung-hong Chan, Po-Chun Chang, Raniere Gaia Costa da Silva, Danilo Dessi, Stefan Dietze, Hajira Jabeen, M. Taimoor Khan, Gabriella Lapesa, David Schoch, and Claudia Wagner. 2024. "The Methods Hub." ESWC (Extended Semantic Web Conference) Greece, 2024, 2024-05-26.
Momeni, Fakhri, and M. Taimoor Khan. 2024. "Transformer-based Multitask Learning German Sexism Detector." KONVENS: Konferenz zur Verarbeitung natürlicher Sprache/Conference on Natural Language Processing, 2024-09-09.
Event
Khan, M. Taimoor, Ulrike Lühe, and Benoît Rihoux. 2025. "2nd MethodsNET conference, The global hub for research methods innovations, excellence and smart use." 2nd MethodsNET conference, The global hub for research methods innovations, excellence and smart use, 2025-09-10 - 2025-09-12.
Momeni, Fakhri, M. Taimoor Khan, Arnim Bleier, and Tony Ross-Hellauer. 2025. "Social Science Meets Web Data: Reproducible and Reusable Computational Approaches (R2CASS)." International AAAI Conference on Web and Social Media (ICWSM) 2025, 2025-06-23 - 2025-06-23.