Atnafu Lambebo Tonja
Scholar

Atnafu Lambebo Tonja

Google Scholar ID: rubyApkAAAAJ
Postdoc at MBZUAI
NLP for low-resource languagesMultilingual language modelsSpeech Technology
Citations & Impact
All-time
Citations
671
 
H-index
16
 
i10-index
22
 
Publications
20
 
Co-authors
9
list available
Resume (English only)
Academic Achievements
  • 1. NAACL 2025: ProverbEval: Exploring LLM Evaluation Challenges for Low-resource Language Understanding
  • 2. NeurIPS 2024 D&B Track: CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark
  • 3. EMNLP 2024: The Zeno’s Paradox of ‘Low-Resource’ Languages and Walia-LLM: Enhancing Amharic-LLaMA by Integrating Task-Specific and Generative Datasets
  • 4. LREC-COLING 2024: EthioLLM: Multilingual Large Language Models for Ethiopian Languages with Task Evaluation
  • 5. NAACL 2024: NLP Progress in Indigenous Latin American Languages
  • 6. EMNLP 2023: The Less the Merrier? Investigating Language Representation in Multilingual Models and Cross-lingual Open-Retrieval Question Answering for African Languages
  • 7. TACL: AfriSpeech-200: Pan-African accented speech dataset for clinical and general domain ASR
  • 8. AACL 2023: MasakhaNEWS
  • 9. INTERSPEECH 2023: AfriNames: Most ASR models 'butcher' African Names
Research Experience
  • Currently a Postdoctoral researcher at Mohamed bin Zayed University of Artificial Intelligence, UAE, working with Prof. Thamar Solorio.
Education
  • PhD in Computer Science from Instituto Politécnico Nacional, Mexico, supervised by Prof. Alexander Gelbukh and Prof. Olga Kolesnikova.
Background
  • Postdoctoral researcher with research interests in NLP for under-resourced languages, multilingual language models, evaluation benchmarks, and speech & multimodal NLP.
Miscellany
  • Social Media: Twitter, LinkedIn, Github, Google Scholar