Pride Kavumba

Pride Kavumba

AI Research Scientist

Biography

I am Pride Kavumba, holding a Ph.D. in Natural Language Processing from Tohoku University, Japan. My research interests include LLMs, evaluation, robustness, interpretability, and common sense reasoning.

Interests
  • Natural Language Processing
  • Evaluation
  • Robustness and Interpretability
  • Knowledge base and Common Sense Reasoning
Education
  • PhD in NLP, 2023

    Tohoku University

  • Master of Information Science (NLP), 2020

    Tohoku University

  • BEng in Electrical Engineering and Electronics Engineering, 2013

    The University of Zambia

Projects

scue
A unified framework for analyzing superficial cues in datasets.

Other Publications

(2023). 因果的 プロンプトによる NLI の敵対的ロバスト性の強化. 言語処理学会第 29 回年次大会.

PDF Cite

(2022). プロンプトモデルは表面的手がかりを 利用するか. 言語処理学会第 28 回年次大会.

PDF Cite

(2021). None the wiser? Adding “None’’Mitigates Superficial Cues in Multiple-Choice Benchmarks. 言語処理学会第 27 回年次大会.

PDF Cite

(2019). Exploring Supervised Learning of Hierarchical Event Embedding with Poincaré Embeddings. 言語処理学会第 25 回年次大会.

PDF Cite

Academic Activities

Reviewer Activities

2024

2023

2022

2021