About
I am a PhD Researcher specializing in Natural Language Processing and Information Retrieval, currently affiliated with the Data Science Group at the University of Innsbruck. I earned my B.Sc. and M.Sc. degrees in Software Engineering from the University of Kurdistan and the University of Isfahan in 2016 and 2019, respectively. During my studies, I was supervised by Dr. Parham Moradi and Prof. Mohammad Ali Nematbakhsh in the Faculty of Computer Engineering.
PhD Researcher
- Birthday: 23 June 1993
- Website: www.jmozafari.com
- City: Shiraz, Iran
- Age: 31
- Degree: Master
- Email: mozafari.jamshid@gmail.com
Publications
Awards
Certificates
Projects
Research Interests
Natural Language Processing
Natural language processing (NLP) is a subfield of linguistics, computer science, and artificial intelligence concerned with the interactions between computers and human language, in particular how to program computers to process and analyze large amounts of natural language data.
Hint Generation
Hint Generation aims to give subtle guidance that leads users toward the correct answer without explicitly revealing it, thereby promoting active learning and cognitive engagement.
Question Answering
Question answering (QA) is a computer science discipline within the fields of information retrieval and natural language processing (NLP), which is concerned with building systems that automatically answer questions posed by humans in a natural language.
Information Retrieval
Information retrieval (IR) is the process of obtaining information system resources that are relevant to an information need from a collection of those resources. Information retrieval is the science of searching for information in a document, searching for documents themselves, and also searching for the metadata that describes data, and for databases of texts, images or sounds.
Machine Reading Comprehension
MRC scans documents and extracts meaning from the text, just like a human reader. You can ask MRC questions about a document and it will use different parts of the content until an answer is formed.
Information Extraction
Information extraction (IE) is the task of automatically extracting structured information from unstructured and/or semi-structured machine-readable documents and other electronically represented sources.
Skills
Deep Learning and Neural Networks
Natural Language Processing
Machine Learning Libraries
Programming Languages
Miscellaneous
Operating Systems
Resume
Research
PhD Researcher
March 2023 - Present
Administrator: Prof. Adam Jatowt
PhD Researcher, Data Science Group, University of Innsbruck
Our goal is to pioneer a new field focused on generating hints rather than providing final answers. This approach encourages users to think critically and reason independently, reducing reliance on large language models (LLMs). Additionally, we aim to introduce innovative evaluation metrics, frameworks, datasets, and methods to support and advance this emerging field.
Research Assistant
Jul 2020 - Jan 2023
Administrator: Prof. Mohammad Ali Nematbakhsh
Research Assistant, BigDATA lab, University of Isfahan
The aim of the Porseman project is to develop an open-domain QA system for the Persian language.
Jul 2019 - Jun 2020
Advisor: Dr. Afsaneh Fatemi
Research Assistant, BigDATA lab, University of Isfahan
We are performing research on the QA systems especially the Answer Selection component using BERT models.
Education
Ph.D. of Computer Science
2023 - 2027
Supervisor: Prof. Adam Jatowt
Data Science Group, University of Innsbruck, Innsbruck, Austria
- Thesis: “Automatic Hint Generation for Questions”
M.Sc. of Software Engineering
2016 - 2019
Supervisor: Prof. Mohammad Ali Nematbakhsh Advisor: Dr. Afsaneh Fatemi
BigDATA Lab, University of Isfahan, Isfahan, Iran
- Thesis: “Factoid Question Answering for Named Entity”: 4.0 / 4.0
- Overall GPA: 4.0 / 4.0
- 3rd Student Award in Computer Engineering
B.Sc. of Software Engineering
2011 - 2016
Supervisor: Dr. Parham Moradi
Department of Computer Engineering, University of Kurdistan, Sanandaj, Iran
- Thesis: “Online Evolution and Training System for Data Structure and Algorithms”: 4.0 / 4.0
- Overall GPA: 3.5 / 4.0
- 3rd Student Award in Computer Engineering
Teaching Experience
University of Innsbruck
Sep 2024 - Feb 2025
Teacher: Prof. Jan Beutel Introduction to Networking and Internet, Proseminar Teacher
March 2024 - Jun 2024
Teacher: Prof. Justus Piater Data Structures and Algorithms, Proseminar Teacher
Sep 2023 - Feb 2024
Teacher: Prof. Jan Beutel Introduction to Networking and Internet, Proseminar Teacher
University of Isfahan
Sep 2017 - Feb 2018
Teacher: Dr. Fatemeh Raji Data Structures and Algorithms, Teacher Assistant
Sep 2016 - Jun 2017
Teacher: Dr. Ahmad Zaeri Advanced programming in JAVA, Teacher Assistant
University of Kurdistan
Sep 2014 - Jun 2015
Teacher: Dr. Parham Moradi Data Structures and Algorithms, Teacher Assistant
Sep 2013 - Jun 2014
Teacher: Dr. Amanj Khoramian Advanced programming in JAVA, Teacher Assistant
Sep 2013 - Jun 2014
Teacher: Dr. Amanj Khoramian Data Structures and Algorithms, Teacher Assistant
Sep 2012 - Jun 2013
Teacher: Dr. Amanj Khoramian Advanced programming in JAVA, Teacher Assistant
Professional Experience
University of Kurdistan
Sep 2012 - Jun 2014
Position: Chairman and Primary member ACM/ICPC Programming Contest
Sep 2012 - Jun 2014
Position: Chairman and Primary member ACM Association
Sep 2012 - Jun 2014
Position: Chairman and Primary member Computer Engineering Association
Publications
30 conference and journal papers, 1 book, plus evaluation-lab organization, manuscripts under review, and preprints.
- All
- Conference
- Journal
- Book
- Evaluation Lab
- Under Review
- Preprint
HintEval: An Open-Source Python Toolkit for Hint Generation and Hint Evaluation
Jamshid Mozafari, Bhawna Piryani, Abdelrahman Abdallah, Adam Jatowt
Empirical Methods in Natural Language Processing: EMNLP Demo 2026, ACL (2026)
DOI: 10.18653/2026.emnlp-demos.2
Output of: DataScience Lab
Question Difficulty Estimation for Large Language Models via Answer Plausibility Scoring
Jamshid Mozafari, Bhawna Piryani, Adam Jatowt
Annual Meeting of the Association for Computational Linguistics: ACL 2026, ACL (2026)
DOI: 10.18653/v1/2026.acl-long.510
Inferential Question Answering
Jamshid Mozafari, Hamed Zamani, Guido Zuccon, Adam Jatowt
International World Wide Web Conference 2026, ACM (2026)
DOI: 10.1145/3774904.3792653
Output of: DataScience Lab
It's High Time: A Survey of Temporal Question Answering
Bhawna Piryani, Abdelrahman Abdallah, Jamshid Mozafari, Avishek Anand, Adam Jatowt
Annual Meeting of the Association for Computational Linguistics: ACL 2026, ACL (2026)
DOI: 10.18653/v1/2026.acl-long.1332
Output of: DataScience Lab
Pretraining Exposure Explains Popularity Judgments in Large Language Models
Jamshid Mozafari, Bhawna Piryani, Adam Jatowt
International SIGIR Conference on Research and Development in Information Retrieval 2026, ACM (2026)
DOI: 10.1145/3805712.3809958
Output of: DataScience Lab
Context Convergence Improves Answering Inferential Questions
Jamshid Mozafari, Bhawna Piryani, Adam Jatowt
International SIGIR Conference on Research and Development in Information Retrieval 2026, ACM (2026)
DOI: 10.1145/3805712.3809848
Output of: DataScience Lab
Rankify: A Comprehensive Python Toolkit for Retrieval, Re-Ranking, and Retrieval-Augmented Generation
Abdelrahman Abdallah, Bhawna Piryani, Jamshid Mozafari, Mohammed Ali, Adam Jatowt
Annual Meeting of the Association for Computational Linguistics: ACL Demo 2026, ACL (2026)
DOI: 10.18653/v1/2026.acl-demo.21
Output of: DataScience Lab
Does the Temporal Focus of Information Affect Confidence Calibration in LLMs? A Study of Temporal Effects on Model Confidence
Bhawna Piryani, Jamshid Mozafari, Adam Jatowt
Findings of the Association for Computational Linguistics: EMNLP 2026, ACL (2026)
DOI: 10.18653/2026.findings-emnlp.1854
Output of: DataScience Lab
HintLab: An Interactive Tool for Hint Generation, Analysis and Evaluation
Gabriel Kofler, Jamshid Mozafari, Adam Jatowt
International Conference on Information and Knowledge Management 2026, ACM (2026)
DOI: 10.1145/3799682.3840251
Evaluating Robustness of LLMs in Question Answering on Multilingual Noisy OCR Data
Bhawna Piryani, Jamshid Mozafari, Abdelrahman Abdallah, Antoine Doucet, Adam Jatowt
International Conference on Information and Knowledge Management 2025, ACM (2025)
DOI: 10.1145/3746252.3761295
Output of: DataScience Lab
RerankArena: A Unified Platform for Evaluating Retrieval, Reranking and RAG with Human and LLM Feedback
Abdelrahman Abdallah, Mahmoud Abdalla, Bhawna Piryani, Jamshid Mozafari, Mohammed Ali, Adam Jatowt
International Conference on Information and Knowledge Management 2025, ACM (2025)
DOI: 10.1145/3746252.3761484
Output of: DataScience Lab
DeAR: Dual-Stage Document Reranking with Reasoning Agents via LLM Distillation
Abdelrahman Abdallah, Jamshid Mozafari, Bhawna Piryani, Adam Jatowt
Findings of the Association for Computational Linguistics: EMNLP 2025, ACL (2025)
DOI: 10.18653/v1/2025.findings-emnlp.306
Output of: DataScience Lab
How Good are LLM-based Rerankers? An Empirical Analysis of State-of-the-Art Reranking Models
Abdelrahman Abdallah, Bhawna Piryani, Jamshid Mozafari, Adam Jatowt
Findings of the Association for Computational Linguistics: EMNLP 2025, ACL (2025)
DOI: 10.18653/v1/2025.findings-emnlp.305
Output of: DataScience Lab
Wrong Answers Can Also Be Useful: PlausibleQA - A Large-Scale QA Dataset with Answer Plausibility Scores
Jamshid Mozafari, Abdelrahman Abdallah, Bhawna Piryani, Adam Jatowt
International SIGIR Conference on Research and Development in Information Retrieval 2025, ACM (2025)
DOI: 10.1145/3726302.3730299
Output of: DataScience Lab
WikiHint: A Human-Annotated Dataset for Hint Ranking and Generation
Jamshid Mozafari, Florian Gerhold, Adam Jatowt
International SIGIR Conference on Research and Development in Information Retrieval 2025, ACM (2025)
DOI: 10.1145/3726302.3730284
Output of: DataScience Lab
ASRank: Zero-Shot Re-Ranking with Answer Scent for Document Retrieval
Abdelrahman Abdallah, Jamshid Mozafari, Bhawna Piryani, Adam Jatowt
Findings of the Association for Computational Linguistics: NAACL 2025, ACL (2025)
DOI: 10.18653/v1/2025.findings-naacl.161
Output of: DataScience Lab
DynRank: Improving Passage Retrieval with Dynamic Zero-Shot Prompting Based on Question Classification
Abdelrahman Abdallah, Jamshid Mozafari, Bhawna Piryani, Mohammed M Abdelgwad, Adam Jatowt
International Conference on Computational Linguistics: COLING 2025, ACL (2025)
ACL Anthology
Output of: DataScience Lab
Navigating the Landscape of Hint Generation Research: From the Past to the Future
Anubhav Jangra, Jamshid Mozafari, Adam Jatowt, Smaranda Muresan
Transactions of the Association for Computational Linguistics (2025) 
DOI: 10.1162/tacl_a_00751
Output of: DataScience Lab
Exploring Hint Generation Approaches in Open-Domain Question Answering
Jamshid Mozafari, Abdelrahman Abdallah, Bhawna Piryani, Adam Jatowt
Findings of the Association for Computational Linguistics: EMNLP 2024, ACL (2024)
DOI: 10.18653/v1/2024.findings-emnlp.546
Output of: DataScience Lab
Detecting Temporal Ambiguity in Questions
Bhawna Piryani, Jamshid Mozafari*, Abdelrahman Abdallah*, Adam Jatowt
Findings of the Association for Computational Linguistics: EMNLP 2024, ACL (2024)
DOI: 10.18653/v1/2024.findings-emnlp.562
Output of: DataScience Lab
TriviaHG: A Dataset for Automatic Hint Generation from Factoid Questions
Jamshid Mozafari, Anubhav Jangra, Adam Jatowt
International SIGIR Conference on Research and Development in Information Retrieval 2024, ACM (2024)
DOI: 10.1145/3626772.3657855
Output of: DataScience Lab
ChroniclingAmericaQA: A Large-scale Question Answering Dataset based on Historical American Newspaper Pages
Bhawna Piryani, Jamshid Mozafari, Adam Jatowt
International SIGIR Conference on Research and Development in Information Retrieval 2024, ACM (2024)
DOI: 10.1145/3626772.3657891
Output of: DataScience Lab
FarsNewsQA: A Deep Learning-based Question Answering System for the Persian News Articles
Arefeh Kazemi, Zahra Zojaji, Mahdi Malverdi, Jamshid Mozafari, Fatemeh Ebrahimi, Negin Abadani, Mohammad Reza Varasteh, Mohammad Ali Nematbakhsh
Information Retrieval Journal (2023) 
DOI: 10.1007/s10791-023-09417-2
Output of: BigDATA Lab
PerAnSel: A Novel Deep Neural Network-based System for Persian Question Answering
Jamshid Mozafari, Arefeh Kazemi, Parham Moradi, Mohammad Ali Nematbakhsh
Computational Intelligence and Neuroscience (2022) 
DOI: 10.1155/2022/3661286
Output of: BigDATA Lab
PersianQuAD: The Native Question Answering Dataset for the Persian Language
Arefeh Kazemi, Jamshid Mozafari, Mohammad Ali Nematbakhsh
IEEE Access (2022) 
DOI: 10.1109/ACCESS.2022.3157289
Output of: BigDATA Lab
ParSQuAD: Machine Translated SQuAD dataset for Persian Question Answering
Negin Abadani, Jamshid Mozafari, Afsaneh Fatemi, Mohammad Ali Nematbakhsh, Arefeh Kazemi
International Conference on Web Research 2021, IEEE (2021)
DOI: 10.1109/ICWR51868.2021.9443126
Output of: BigDATA Lab
ParSQuAD: Persian Question Answering Dataset based on Machine Translation of SQuAD 2.0
Negin Abadani, Jamshid Mozafari, Afsaneh Fatemi, Mohammad Ali Nematbakhsh, Arefeh Kazemi
International Journal on Web Research (2021)
DOI: 10.22133/ijwr.2021.293313.1101
Output of: BigDATA Lab
BAS: An Answer Selection Method Using BERT Language Model
Jamshid Mozafari, Afsaneh Fatemi, Mohammad Ali Nematbakhsh
Journal of Computing and Security (2021)
DOI: 10.22108/jcs.2021.128002.1066
Output of: BigDATA Lab
A Method For Answer Selection Using DistilBERT And Important Words
Jamshid Mozafari, Afsaneh Fatemi, Parham Moradi
International Conference on Web Research 2020, IEEE (2020)
DOI: 10.1109/ICWR49608.2020.9122302
Output of: BigDATA Lab
Improved Answer Selection for Factoid Questions
Jamshid Mozafari, Mohammad Ali Nematbakhsh, Afsaneh Fatemi
International Conference on Computer and Knowledge Engineering 2019, IEEE (2019)
DOI: 10.1109/ICCKE48569.2019.8965131
Output of: M.Sc. Thesis
Fundamentals of Data Structures (with ACM Problems)
Fatemeh Raji, Jamshid Mozafari
Iranian Student Book Agency (2019)
ISBN: 978-600-318-238-7
TemporalWebLab at CLEF 2027: Evaluating Recency and Change across the Web
Bhawna Piryani, Jamshid Mozafari, Sawood Alam, Ricardo Campos, Avishek Anand, Adam Jatowt
Conference and Labs of the Evaluation Forum (CLEF 2027) · Accepted evaluation lab (not a proceedings paper) (2027)
Official website
How Training-Data Exposure Relates to Factual Behavior in LLMs: A Large-Scale Empirical Study
Jamshid Mozafari, Bhawna Piryani, Zahra Sinaei, Maryam Keshvari, Jie Ni, Adam Jatowt
Submitted to Annual Meeting of the Association for Computational Linguistics: ACL 2027, ACL · Under review (not accepted) (2027) 
How Pretraining Evidence Relates to LLM Reliance on External Context
Jamshid Mozafari, Maryam Keshvari, Bhawna Piryani, Zahra Sinaei, Adam Jatowt
Submitted to European Conference on Information Retrieval: ECIR 2027, Springer · Under review (not accepted) (2027)
Evidence Monotonicity in Retrieval Scoring
Jamshid Mozafari, Bhawna Piryani, Maryam Keshvari, Zahra Sinaei, Adam Jatowt
Submitted to European Conference on Information Retrieval: ECIR 2027, Springer · Under review (not accepted) (2027)
PARSE: An Open-Domain Reasoning Question Answering Benchmark for Persian
Jamshid Mozafari, Seyed Parsa Mousavinasab, Adam Jatowt
arXiv preprint (2026)
DOI: 10.48550/arXiv.2602.01246
Output of: DataScience Lab
From Retrieval to Generation: Comparing Different Approaches
Abdelrahman Abdallah, Jamshid Mozafari, Bhawna Piryani, Mohammed Ali, Adam Jatowt
arXiv preprint (2025)
DOI: 10.48550/arXiv.2502.20245
Attention-based Pairwise Multi-Perspective Convolutional Neural Network for Answer Selection in Question Answering
Jamshid Mozafari, Afsaneh Fatemi, Mohammad Ali Nematbakhsh
arXiv preprint (2019)
DOI: 10.48550/arXiv.1909.01059
Output of: M.Sc. Thesis
Awards
Certificates
Projects
Porseman
In this project, we are working on a QA system and developing it for the Persian language. We started working on this system from scratch.
KOJ (Kurdistan Online Judge)
It's an online (ASP.NET) and offline (JAVA) judger for ACM/ICPC problems. It has been implemented for judging Design and Analysis of Algorithms and Data Structure practices and has been used in the University Of Kurdistan for these courses.
Persia Search Engine
It is a search engine named Persia that is written in Java with Apache Lucene library. It contains some interesting customization settings; for example, it fetches random documents from Wikipedia and indexes them. You can also switch searching language between Persian and English.
Voice Chat room
It's a Chat room that is written in Java. It contains some interesting features. For example, you can have voice chat and text chat. You can also block every member of the chat room. There are many emojis in this chat room. I've used Socket programming in this project.
Graph Analysis on Tarafdari.com
It's an implementation for the "Analysis of relationship between members of Tarafdari.com" publication. I've used NetworkX and BeautifulSoup library in this project.
References
Jamshid is a diligent and disciplined individual and has shown to me a great interest in perusing higher education with qualities that have made his motivated and determined for bypassing impediments to achieve great places ...
Mohammad Ali Nematbakhsh
Professor, University of Isfahan
I was totally aware of Jamshid’s entrepreneurship mindset and his solid background in programming and computer science, so I asked him to join my company. He is very personable and easy to work with great team-working skills ...
Parham Moradi
Associate Professor, University of RMIT
Jamshid is a deep thinker who reads a lot and takes any opportunity to learn something. He has a curious and investigative personality; I believe that he has the strongest motivation and potential to be an outstanding innovative ...
Afsaneh Fatemi
Associate Professor, University of Isfahan
In my view, Jamshid is a well-educated, precise, persistent, and hard-working person in solving practical problems. On the personal side, he is a very polite and brilliant person with great communication skills and pleasant behavior ...
Fatemeh Raji
Senior Lecturer, University of De Montfort
Jamshid has an outstanding academic profile. He has great potential in research. Serving as my research assistant, he was able to successfully complete all research tasks that I assigned to him. Jamshid has excellent knowledge of deep learning ...
Arefeh Kazemi
Research Fellow, University College Dublin
Contact
Location:
Innsbruck, Tyrol State, Austria
Email:
jamshid.mozafari@uibk.ac.at
mozafari.jamshid@gmail.com