Jean de Dieu Nyandwi

I am a visiting researcher at CMU's Language Technologies Institute (NeuLab), working with Prof. Graham Neubig and collaborators across CMU. I have also had the privilege of collaborating with Prof. Deva Ramanan. I work on machine learning and multimodal AI, with a focus on understanding when and why models fail and developing methods to diagnose and mitigate these failures. I hold an MS in Engineering AI from Carnegie Mellon University.

Publications

TimeBlind: A Spatio-Temporal Compositionality Benchmark for Video LLMs
Baiqi Li, Kangyi Zhao, Ce Zhang, Chancharik Mitra, Jean de Dieu Nyandwi, Gedas Bertasius
arXiv 2026
ArXiv · Page

Grounding Multilingual Multimodal LLMs With Cultural Knowledge
Jean de Dieu Nyandwi, Yueqi Song, Simran Khanuja, Graham Neubig
EMNLP 2025 (Oral)
ArXiv · Page · Dataset · Models

Pangea: A Fully Open Multilingual Multimodal LLM for 39 Languages
Xiang Yue*, Yueqi Song*, Akari Asai, Seungone Kim, Jean de Dieu Nyandwi, Simran Khanuja, Anjali Kantharuban, Lintang Sutawika, Sathyanarayanan Ramamoorthy, Graham Neubig
ICLR 2025
ArXiv · Page · Models · Code

NaturalBench: Evaluating Vision-Language Models on Natural Adversarial Samples
Baiqi Li*, Zhiqiu Lin*, Wenxuan Peng*, Jean de Dieu Nyandwi*, Daniel Jiang, Zixian Ma, Simran Khanuja, Ranjay Krishna, Graham Neubig, Deva Ramanan
NeurIPS 2024
ArXiv · Page · Benchmark

Featured Projects

Complete Machine Learning Package — Comprehensive collection of 30+ notebooks on Python, data analysis, classical ML, computer vision, and NLP. Used by thousands of learners.
Web · GitHub

Deep Revision — Deep technical dives into ML research and ideas.
deeprevision.github.io

ModernConvNets — Designs, implementations, and annotated papers of 13 modern CNN architectures (AlexNet through ConvNeXt).
GitHub

Videos — Lectures on transformer architecture and vision transformers.
Transformer Architecture · Vision Transformers

News

Fun Facts