Hi, I’m Adiba Ibnat Hossain

I am a PhD student in Computer Science at Northern Illinois University, specializing in Natural Language Processing, Machine Learning, and Data Science. My current research focuses on hallucination in large language models, a critical challenge for developing trustworthy generative AI systems. I am passionate about solving text-driven problems with advanced language models and exploring innovative technologies that bridge the gap between academic research and real-world applications.

Education

Northern Illinois University
PhD in Computer Science (August 2024 - Present)

Chittagong University of Engineering & Technology (CUET)
B.Sc. in Computer Science and Engineering (August 2022)

Work Experience

Research Assistant
Summer 2025
DATA Laboratory
Department of Computer Science,
Northern Illinois University, Dekalb, Illinois

Teaching Assistant
Fall 2024, Spring 2025, Fall 2025
Department of Computer Science,
Northern Illinois University, Dekalb, Illinois

Lecturer on Study Leave (Dec 4, 2022– Present)
Department of Computer Science and Engineering,
Premier University, Chattogram, Bangladesh

Lecturer (September 5, 2022– December 2, 2022)
Department of Computer Science and Engineering,
Southern University Bangladesh, Chattogram, Bangladesh

Publications

• Adiba Ibnat Hossain, Sagnik Ray Choudhury and Hamed Alhoori, ``SciHallu: A Multi-Granularity Hallucination Detection Dataset for Scientific Writing”, Accepted in International Joint Conference on Natural Language Processing & Asia-Pacific Chapter of the Association for Computational Linguistics, 2025 (IJCNLP-AACL-2025) Link

• Adiba Ibnat Hossain, Miftahul Jannat Mokarrama and Hamed Alhoori , ``Quantifying Contextual Hallucinations in NLP Research Papers Before and After the LLM Era”, Accepted in 2nd International Workshop on Artificial Intelligence for the Science of Science (AI4SciSci'25) colocated with JCDL'25.

• Adiba Ibnat Hossain and Asaduzzaman, ``Bangla Text Search Engine Using Pointwise Approach of Learn to Rank(Ltr) Algorithm”, In 2022 IEEE International Women in Engineering (WIE) Conference on Electrical and Computer Engineering (WIECON-ECE), pages 227–232, 2022. Link

• Adiba Ibnat Hossain, Sabitri Sikder, Annesha Das, and Ashim Dey, ``Applying Machine Learning Classifiers on ECG Dataset for Predicting Heart Disease”, In 2021 International Conference on Automation, Control and Mechatronics for Industry 4.0 (ACMI), pages 1–6, 2021. Link

Poster Presentation

• Adiba Ibnat Hossain, Hamed Alhoori, "From Pre-LLM to Post-LLM: A Quantitative Study of Hallucination in Research Papers Before and After the LLM Era", Midwest Speech and Language Days 2025 (MSLD 2025)

• Adiba Ibnat Hossain, Md Sabir Hossain, "Freelancer Assistance Using Job Sequencing Algorithm", International Conference on Natural Science & Technology(ICNST 2019). Published in the book of abstracts.

Award

Research Mentorship Training Award-2025
Got selected for a competitive research mentorship training program and awarded a $500 stipend
Organizer: Office of Student Engagement and Experiential Learning, Northern Illinois University

Diversity and Inclusion Subsidy Awards at IJCNLP-AACL 2025
Registration Fee and ACL Membership Fee Waiver

Skills

✦ Programming Languages : C, C++, Java, Python
✦ Web Development : HTML, CSS, PHP, Javascript, MySQL, Firebase, Flask
✦ Database : MySQL
✦ Frameworks : React Native, ReactJS
✦ ML/NLP Libraries : BeautifulSoup, NLTK, BNLTK, scikit-learn, matplotlib, pandas
✦ Multi-agent Framework : Autogen

Academic Projects

◯AutoHall: A Hallucination-aware Autonomous Multi-agent System (In progress)
• An autonomous multi-agent system that autonomously designs the agents, divides a task into subtasks, checks hallucination in the subtask result through feedback loop, modifies the discrepencies and passes the subtask result to the next agent. The aim of this work is to reduce hallucination propagation among the agents.
Framework: Autogen, Language: Python, IDE: VS code

◯SciHallu: A Multi-Granularity Hallucination Detection Dataset for Scientific Writing
• A dataset for identifying hallucinations in academic text at three levels of granularity: token, sentence, and paragraph.
Language: Python, IDE: VS code

◯Quantifying Contextual Hallucinations in NLP Research Papers Before and After the LLM Era
• Detected and analyzed context inconsistency, a type of faithfulness hallucination in research papers across two distinc periods: before the advent of LLMs and after their integration.
Language: Python, IDE: VS code

◯ Research Paper Subject Category Classification Using Traditional Classifiers and LLM
• Created a novel dataset and built some classification models to perform a comparative analysis among the models.
Language: Python, Platform: Google Colab

◯ A Bangla Text Search Engine Using Pointwise Approach of Learn to Rank(LtR) Algorithm
• A web application that provides search result of a user query in Bangla in a particular ranked order. This project is a combination of ML, NLP, Web crawling. Applied the Point-wise Approach of Learn to Rank Algorithm instead of using the conventional BM25 function.
Language: Python, HTML, CSS, DB: MySQL IDE: PyCharm Others: Flask API, Google Colab

◯ Heart Disease Prediction Using Machine Learning Algorithms
• Applied machine learning classifiers to implement a heart disease prediction model using ECG dataset.

◯ Website for High School Science Contents
• Awebsite for high school students who can easily get guideline to learn science topics with the help of the uploaded contents and interact through the comment section.
Language: HTML, CSS, PHP, DB: MySQL Platforms: Atom, XAMPP

◯ Azheimer’s Disease Prediction App (Android Application)
• An android app that predicts the existence of alzheimer’s disease based on user’s symptoms. IDE: VS code, Jupyter Notebook Framework: React Native DB: Firebase Emergency Alarm for Elderly People (Android Application) • An android app that gives alarm to the favourite contacts of an elderly person by means of shake detection.
Language: Java, IDE: Android Studio

◯ Emergency Alarm for Elderly People (Android Application)
• An android app that gives alarm to the favourite contacts of an elderly person by means of shake detection.
Language: Java, IDE: Android Studio

Supervised Projects

◯ CONFER: A Blood Donation Recommendation System

◯ Leveraging Transformer Language Models for Title Generation in Scientific Articles

◯ Insights into Opinion Mining in Bengali: A Deep Learning Approach to Multi-class Sentiment Analysis of Facebook Comments

◯ A Website to Facilitate Driver Hiring Process