cv

Research Assistant

Digital Science Center (DiSC), University of Innsbruck

Research assistant working on NLP and information retrieval within the Digital Science Center.

• Knowledge extraction and information retrieval from unstructured text documents.
• Methods for natural language processing and information retrieval.
• Application of text mining methods to the field of digital history.

Machine Learning Researcher

Università Ca' Foscari

Researcher working on ML models for climate change and risk assessment.

• Comparative analysis of graph neural networks and random forests for climate-related tasks.
• Contributed to a review paper on ML/AI models for risk assessments.
• Surveyed graph neural networks for spatio-temporal data.

Machine Learning Engineer

KMG Engineering

Machine learning engineer focusing on vision and NLP applications.

• Developed GAN-based models for image inpainting.
• Built English grammar correction models using deep learning.
• Worked on curve detection and tracking tasks.

Machine Learning Engineer (part-time)

DISCO App

Machine learning engineer working on OCR and NLP for digital receipts.

• Worked on receipt extraction, OCR systems, and NLP components.
• Built and improved OCR accuracy for receipts and downstream information extraction and classification.
• Contributed to an application deployed on the Google Play Store.

Machine Learning Researcher

National Open Research Laboratory for Information and Space Technologies, Satbayev University

Researcher focusing on handwriting recognition, table detection, and document analysis.

• Built handwritten Kazakh and Russian databases for handwriting recognition research.
• Reviewed recent approaches to handwritten recognition for Cyrillic characters.
• Created a table detection dataset and developed models for table detection and classification.

Software Developer

CCC at Limkokwing University

Software developer working on web applications with database backends.

• Developed and implemented a scanning component using PHP and MySQL.
• Designed databases and table structures using n-tier architecture for web applications.
• Built web applications using the ExpressionEngine PHP framework.

Research and Teaching Assistant

Assiut University, Faculty of Computers and Information

Research and teaching assistant in Computer Science.

• Taught classes, supervised laboratory sessions, and graded assignments and projects.
• Represented teams in meetings with executives to discuss project goals and milestones.
• Kept up with emerging technologies and applied them to teaching and research projects.

Web Developer

FastKood Company

Web developer building data-driven websites and internal tools.

• Converted UI mockups into HTML, JavaScript, AJAX, and JSON.
• Worked with UNIX and Apache servers.
• Developed data architecture designs for targeted customer analysis.
• Created workflow charts and diagrams to support production teams and meet client deadlines.

Software Developer

Overcoffeesolutions

Software developer focusing on object-oriented applications.

• Developed object-oriented software and intuitive graphical user interfaces.
• Implemented scanning components using MySQL and solid database design.
• Built multiple web applications following n-tier architecture.

Reviewer

Conference and Journal Reviewer

Reviewer for major conferences and journals in NLP, IR, and computer vision.

• Conference reviewer: ACL, SIGIR, COLING, EMNLP, LREC-COLING, WACV.
• Journal reviewer: Pattern Recognition Letters, IET NBT, Heliyon, IET Signal Processing.

PhD

University of Innsbruck

Computer Science

MSc

Satbayev University, Faculty of Information and Telecommunication Technologies

Data Science and Machine Learning

Pre-Master

Assiut University, Faculty of Computer and Information

Computer Science

BSc

Assiut University, Faculty of Computer and Information

Computer Science

Scholarship for Master's Studies

Satbayev University

Scholarship to study for a Master's degree in Data Science and Machine Learning at Satbayev University.

Evaluating Temporal Robustness of Large Language Models

ACL Findings 2025

Evaluation of how LLM performance changes under temporal shifts in data.

TempDPR: Temporal Dense Passage Retrieval for Explicit Temporal Questions

Preprint

Temporal dense passage retrieval model for time-sensitive questions.

From Retrieval to Generation: Evaluating the Best Approach

Preprint

Study comparing retrieval-based and generation-based approaches for QA and related tasks.

CascadePLS-ViT: Cascade With Patch-Level Self-Supervised Vision Transformers for Breast Cancer Classification in Mammography

ISBI 2025

Patch-level self-supervised ViT cascade for mammographic breast cancer classification.

ASRank: Zero-Shot Re-Ranking with Answer Scent for Document Retrieval

NAACL 2025

Zero-shot LLM-based reranking approach that uses answer scent signals.

DynRank: Improving Passage Retrieval with Dynamic Zero-Shot Prompting Based on Question Classification

COLING 2025

Dynamic prompting strategy for better zero-shot passage retrieval using LLMs.

Wrong Answers Can Also Be Useful: PlausibleQA - A Large-Scale QA Dataset with Answer Plausibility Scores

SIGIR 2025

QA dataset with plausibility scores for both correct and incorrect answers.

MultiOCR-QA: Dataset for Evaluating Robustness of LLMs in Question Answering on Multilingual OCR Texts

CIKM 2025

Dataset designed to test robustness of QA models on noisy multilingual OCR outputs.

RankArena: A Unified Platform for Evaluating Retrieval, Reranking and RAG

CIKM 2025

Unified platform for benchmarking retrieval, reranking, and RAG pipelines with human and LLM feedback.

ComplexTempQA: A Large-Scale Dataset for Complex Temporal Question Answering

EMNLP 2025

Large-scale dataset for temporally complex question answering over dynamic collections.

How Good are LLM-based Rerankers? An Empirical Analysis of State-of-the-Art Reranking Models

EMNLP Findings 2025

Empirical analysis of modern LLM-based rerankers across multiple datasets and settings.

DeAR: Dual-Stage Document Reranking with Reasoning Agents via LLM Distillation

EMNLP Findings 2025

Dual-stage LLM-based document reranking framework using reasoning agents and distillation.

ArabicaQA: A Comprehensive Dataset for Arabic Question Answering

SIGIR 2024

Large-scale dataset for Arabic open-domain question answering.

IHRRB-DINO: Identifying High-Risk Regions of Breast Masses in Mammogram Images Using Data-Driven Instance Noise (DINO)

MICCAI 2024

DINO-based framework to highlight high-risk regions in mammogram images.

Detecting Temporal Ambiguity in Questions

EMNLP Findings 2024

Detection of temporally ambiguous questions in open-domain QA datasets.

HiGenQA: Exploring Hint Generation Approaches for Open Domain Question Answering

EMNLP Findings 2024

Study of automatic hint generation strategies for improving open-domain QA.

A Survey of Recent Approaches to Form Understanding in Scanned Documents

Artificial Intelligence Review

Survey of transformers and language models for form understanding and scanned document analysis.

Exploring the State of the Art in Legal QA Systems

Journal of Big Data

Comprehensive review of legal question answering systems and datasets.

Rankify

Creator and maintainer of Rankify, a Python toolkit for retrieval, reranking, and RAG evaluation.

• Comprehensive Python package for information retrieval and reranking evaluation.
• Supports evaluation of retrieval, reranking, and RAG systems with automated metrics and human feedback.
• Integrated with the RankArena platform and has 500+ GitHub stars.

RankArena

Lead developer of RankArena, a unified web platform for evaluating retrieval, reranking, and RAG systems.

• Provides standardized protocols for benchmarking IR models against strong baselines.
• Supports both human and LLM-based feedback for evaluation.
• Accepted at CIKM 2025.