Elia Alghazal

SHARP-RAG: Self-Correcting Hierarchical Agentic Retrieval-Augmented Generation for Multi-Hop Question Answering

Elia Alghazal · Independent Researcher · Beirut, Lebanon

Multi-hop question answering requires chaining evidence across several documents, a setting in which naive RAG frequently fails because it retrieves once and never verifies whether the retrieved context supports an answer. SHARP-RAG addresses this with a four-agent LangGraph pipeline: a Planner, Retriever, Critic, and Synthesizer cooperate in a cyclic stateful graph where the Critic emits a structured JSON verdict that gates answer generation and drives targeted re-retrieval. Evaluated on 20 HotpotQA fullwiki questions, the work's central finding is that critique model calibration determines whether the self-correction loop helps or hurts, more than the architecture itself.

Core finding: critique model calibration, not architecture, determines whether self-correction helps or hurts performance.

CrashLens: Smart Crash Detection and Emergency Response via IoT and Artificial Intelligence

E. Alghazal, G. Khayat, W. Ishak, B. Farhat, M. Allaw · Advised by Dr. C. Boustany, AUST

CrashLens is an edge AI pipeline for automatic vehicle crash detection and emergency dispatch. A Raspberry Pi 5 equipped with IMU, GPS, camera, and a 4G module performs sensor fusion and YOLO inference in real time at the edge. On crash detection, the device packages video, location, and sensor data and routes it via 4G to role-based dashboards for drivers, first responders, and insurance providers, with sub-30-second end-to-end latency. The system includes license plate extraction, an analytics pipeline, and companion mobile applications.