Skip to Main Content

Research

My core research interest is in Deep Learning, particularly the fundamental capabilities of Foundation Models. I aim to understand the limitations of Large Language Models (LLMs), especially in information retrieval and long-context execution, and develop methods for their improvement. At the application level, I leverage these techniques in high-impact, interdisciplinary domains like Earth Observation, Healthcare, and Space Research and Exploration.

Research Focus

Understanding the fundamental capabilities and limitations of Foundation Models and their applications across high-impact domains.

Foundation Models & Information Retrieval

Understanding LLM limitations and developing novel retrieval methods using anomalous pattern detection

Space Missions & Exploration

Developing AI systems for spacecraft monitoring, mission planning, and space exploration technologies

High-Impact Applications

Applying Foundation Models to real-world challenges in Earth observation, healthcare, and neuroscience

Zero-Shot Neural Priors for Generalizable Cross-Subject and Cross-Task EEG Decoding

Zero-Shot Neural Priors for Generalizable Cross-Subject and Cross-Task EEG Decoding

Baimam Boukar JJ, Brandone Fonya, Nchofon Tagha Ghogomu, Pauline Nyaboe, Kipngeno Koech

NeurIPS 2026

EEGBrain-Computer InterfacesZero-Shot LearningNeural DecodingOngoing
Retrieval with Multiple Query Vectors through Anomalous Pattern Detection

Retrieval with Multiple Query Vectors through Anomalous Pattern Detection

Baimam Boukar JJ, Allassan Nken, Miriam Rateike, Celia Cintas

AAAI 2026 Workshop on New Frontiers In Information Retrieval

Information RetrievalDeepScanLLMs EmbeddingsAnomalous PatternReview
Mapping Socioeconomic Air Quality Disparities In Rwanda Using Sentinel-5P TROPOMI Data In Google Earth Engine

Mapping Socioeconomic Air Quality Disparities In Rwanda Using Sentinel-5P TROPOMI Data In Google Earth Engine

Baimam Boukar JJ, Kamikazi Raissa, Bertin Ndahayo, Evelyne Umubyeyi

IEEE MIGARS 2025

Air QualityRemote SensingGoogle EEPublished
Explainable Deep-Learning Based Potentially Hazardous Asteroids Classification Using Graph Neural Networks

Explainable Deep-Learning Based Potentially Hazardous Asteroids Classification Using Graph Neural Networks

Baimam Boukar JJ, Prof. Clarence Worrel

Harvard AstroAI Workshop 2025.

SpaceAIAstrophysicsPresented
Computer Vision-based Calibration of Visual Landing Aids Using Autonomous Drones (PAPI Case Study)

Computer Vision-based Calibration of Visual Landing Aids Using Autonomous Drones (PAPI Case Study)

Baimam Boukar JJ, Alice Mugengano, Jonathan Kayizzi, Richard Muhirwa

IEEE Aerospace Conference 2026.

Computer VisionAIAerospaceReview
Humidity Inference with Geographic Features and Machine Learning for Enhanced Contrail Prediction for African Airspace

Humidity Inference with Geographic Features and Machine Learning for Enhanced Contrail Prediction for African Airspace

Baimam Boukar JJ, Alice Mugengano, Jonathan Kayizzi

IEEE MIGARS 2025

AIAviationContrailsPublished