1. 'From Surveys to Narratives: Rethinking Cultural Value Adaptation in LLMs' accepted to EMNLP 2025. 2. 'NusaAksara: A Multimodal and Multilingual Benchmark for Preserving Indonesian Indigenous Scripts' accepted to ACL 2025. 3. Four papers accepted to EMNLP 2024. 4. 'Towards Measuring and Modeling 'Culture' in LLMs: A Survey' accepted to C3NLP (ACL 2024).
Research Experience
1. University of Edinburgh - Current, focusing on adaptive memory compression and multilinguality. 2. MBZUAI (Abu Dhabi) - Former Research Assistant, working on multilingual, multicultural, and parameter-efficient NLP. 3. UKP Lab, TU Darmstadt - Visiting Researcher, working on parameter-efficient modules.
Education
Degree: PhD; School: University of Edinburgh; Advisors: Edoardo Ponti, Alexandra Birch; Time: Current; Major: Computational Linguistics.
Background
Research Interests: Designing language models that are both compute-efficient and multilingual, so they can reason better under tight budgets and across languages. Professional Field: Improving the efficiency of neural models, with a particular interest in methods that could enable tokenization-free and more compute-optimal architectures. Background: PhD student at the University of Edinburgh, advised by Edoardo Ponti and Alexandra Birch.
Miscellany
Personal Interests: Providing support and help to researchers from underrepresented groups.