Mohab Elkaref
Scholar

Mohab Elkaref

Google Scholar ID: e0dPOagAAAAJ
Research Scientist, IBM Research UK
Natural Language ProcessingDeep Learning
Citations & Impact
All-time
Citations
67
 
H-index
4
 
i10-index
1
 
Publications
20
 
Co-authors
3
list available
Contact
No contact links provided.
Resume (English only)
Academic Achievements
  • Verifiable Chemical Reasoning through Tool-Calling Agentic Workflow, Gabrielle Gaudeau, Shinnosuke Tanaka et al., NeurIPS 2025
  • QGen Studio: An Adaptive Question-Answer Generation, Training and Evaluation Platform, Movina Moses, Mohab Elkaref et al., AAAI 2025
  • NLPeople at L+M-24 Shared Task: An Ensembled Approach for Molecule Captioning from SMILES, Shinnosuke Tanaka, Carol Mak et al., ACL 2024
  • NLPeople at TextGraphs-17 Shared Task: Chain of Thought Questioning to Elicit Decompositional Reasoning, Movina Moses, Vishnudev Kuruvanthodi et al., ACL 2024
  • KnowledgeHub: An end-to-end Tool for Assisted Scientific Discovery, Shinnosuke Tanaka, James Barry et al., IJCAI 2024
  • NLPeople at NADI 2023 Shared Task: Arabic Dialect Identification with Augmented Context and Multi-Stage Tuning, Mohab Elkaref, Movina Moses et al., EMNLP 2023
  • NLPeople at SemEval-2023 Task 2: A Staged Approach for Multilingual Named Entity Recognition, Mohab Elkaref, Nathan Herr et al., ACL 2023
  • Automatic, physical data extraction from scientific publications for application to generative molecular design in computational materials discovery, Ronaldo Giro, Mohab Elkaref et al., APS March Meeting 2023
Research Experience
  • Since joining IBM, his research covers a broader range of Natural Language Processing, Understanding, and Generation tasks, including Named Entity Recognition, Relation Extraction, and Question Answering, as well as deep learning techniques that improve and/or bridge these tasks.
Education
  • PhD from the University of Birmingham, where his research revolved around Dependency Parsing and Deep Learning.
Background
  • A Staff Research Scientist focusing on Natural Language Processing. His research interests include Dependency Parsing, Deep Learning, Named Entity Recognition, Relation Extraction, and Question Answering.
Miscellany
  • Top collaborators include Geeth de Mel, Shinnosuke Tanaka, Movina Moses, and James Barry. Work location is at IBM Research Europe - United Kingdom, Daresbury, England.