# Research Papers By Hassan Jaber

This page summarizes selected research papers, preprints, and academic reports. These works should not be described as peer reviewed unless a specific venue or publication status is provided.

## EmbodimentSemantic
- Title: EmbodimentSemantic: A Spatial Scene-Graph Dataset and Benchmark for Vision-Language Models on Embodied Manipulation Trajectories
- Year: 2026
- Status: arXiv preprint
- Authors: Hassan Jaber, Refinath S N, Luca Cagliero, Christopher E. Mower, Haitham Bou-Ammar
- Link: https://arxiv.org/abs/2607.00020
- Summary: Introduces a spatial scene-graph dataset and benchmark for evaluating how vision-language models understand object relations, viewpoint-dependent structure, and embodied manipulation trajectories in robotics.

## Sparse Fine Tuning In Federated Learning
- Title: Exploring Sparse Fine Tuning in Federated Learning under Data Heterogeneity Constraints
- Year: 2025
- Status: academic report
- Authors: Hassan Jaber, Abedal Salam Al Ashi Abou Shoushe, Hassan Maatouk, Henrique Pedro Rochel
- Link: https://drive.google.com/file/d/1C_q5jGiBj2sFm52YWX5m61wO_9bF-JUT/view?usp=sharing
- Summary: Investigates sparse fine-tuning in federated learning under distributed data heterogeneity, with attention to model performance and communication efficiency.

## Speaker Age Prediction
- Title: Predicting Speaker Age from Acoustic and Linguistic Features
- Year: 2024
- Status: academic report
- Authors: Hassan Jaber, Abedal Salam Al Ashi Abou Shoushe
- Link: https://drive.google.com/file/d/1t3Bu12mT1G68Bmiotu9wnEfNH7d1n8Yx/view?usp=sharing
- Summary: Estimates speaker age using acoustic and linguistic features, including MFCCs, spectral data, pitch, and a tuned XGBoost regressor.

## COVID-19 Detection With Federated Learning
- Title: COVID-19 Detection using Federated Learning and Trust Algorithm
- Year: 2022
- Status: academic report
- Authors: Hassan Jaber
- Link: https://drive.google.com/file/d/17gbJMMRO4vjb8gslZ6NIAMtkENeinJXw/view?usp=sharing
- Summary: Explores federated learning for COVID-19 chest X-ray detection with a trust-aware approach for detecting data-poisoned clients.