Loading portfolio content
Amir Mohammad Sharafaddini
I study large language models (LLMs), with a focus on AI agents and reasoning. I’m interested in how models break down complex problems, decide when to use tools, and revise their plans when new evidence changes the answer. My work connects these research questions with building and evaluating agents that can carry out multi-step tasks reliably.
Background
Education
M.Sc. in Computer Science
Tarbiat Modares University
Tehran, Iran
- Expected graduation: Fall 2028.
B.Sc. in Computer Science
Shahid Bahonar University of Kerman
Kerman, Iran
- Research supervisor: Dr. Najme Mansouri.
- Neural Networks — 20/20
- Data Mining — 20/20
- Topics in Computer Science — 20/20
- Data Structures and Algorithms — 19.75/20
- Genetic Algorithms — 20/20
- Design and Analysis of Algorithms — 20/20
- Combinatorics — 19.75/20
Research interests
- LLM Evaluation and Reliability
- Agentic AI
- Interpretability
- System 2 AI and reasoning
Experience
Teaching, Research & Workshops
Teaching Assistant — Fall 2026
Sharif University of Technology
Tehran, Iran
- Mobile Programming — Dr. Ali Najimi.
- Compiler Design — Dr. Samane Hosseinmardi.
- Machine Learning Systems Engineering — Dr. Maryam Ramezani.
Teaching Assistant — Fall 2026
Shahid Bahonar University of Kerman
Kerman, Iran
- Data Mining (graduate course) — Dr. Najme Mansouri.
- Data Structures — Dr. Najme Mansouri.
- Advanced Programming — Dr. Reyhane Ghafari.
- Introduction to Data Mining — Dr. Behnam Mohammad Hassani Zadeh.
- Modeling in Occupational Health and the Work Environment — Dr. Reyhane Ghafari.
Organizer and Instructor — Deep Learning and the Rise of Large Language Models — Summer 2026
Iran's National Elites Foundation, Kerman Province Branch
Kerman, Iran
- Workshop for Ph.D. students, Summer 2026.
Teaching Assistant — Spring 2026
Shahid Bahonar University of Kerman
Kerman, Iran
- Fuzzy Logic — Dr. Reyhane Ghafari.
- Data Structures and Algorithms — Dr. Najme Mansouri.
- Neural Networks — Dr. Behnam Mohammad Hassani Zadeh.
Teaching Assistant — Fall 2025
Shahid Bahonar University of Kerman and Sharif University of Technology
Kerman and Tehran, Iran
- Machine Learning, Sharif University of Technology — Dr. Ali Sharifi-Zarchi.
- Fundamentals of Programming (Python), Sharif University of Technology — Dr. Mohammad Nazari.
- Topics in Data Science (graduate course), Shahid Bahonar University of Kerman — Dr. Najme Mansouri.
Organizer and Instructor — From Machine Learning to Deep Neural Networks — Summer 2025
Iran's National Elites Foundation, Kerman Province Branch
Kerman, Iran
- Workshop for Ph.D. students, Summer 2025.
Teaching Assistant — Spring 2025
Shahid Bahonar University of Kerman and Sharif University of Technology
Kerman and Tehran, Iran
- Advanced Programming (Python), Sharif University of Technology — Dr. Reyhane Ghafari.
- Fundamentals of Programming (Python), Sharif University of Technology — Dr. Marzieh Sadri.
- Data Structures and Algorithms — Dr. Najme Mansouri.
- Data Science Algorithms — Dr. Najme Mansouri.
Teaching Assistant — Fall 2024
Shahid Bahonar University of Kerman
Kerman, Iran
- Advanced Programming (C++) — Dr. Behnam Mohammad Hassani Zadeh.
- Data Mining (graduate course) — Dr. Najme Mansouri.
- Topics in Data Science (graduate course) — Dr. Najme Mansouri.
Teaching Assistant — Spring 2024
Shahid Bahonar University of Kerman
Kerman, Iran
- Advanced Programming (C++) — Dr. Aboozar Zandvakili.
- Topics in Soft Computing and Artificial Intelligence (graduate course) — Dr. Najme Mansouri.
Organizer and Instructor — Deep Learning — Spring 2024
Shahid Bahonar University of Kerman
Kerman, Iran
- Graduate workshop, Spring 2024.
Organizer and Instructor — Python and Machine Learning — Fall 2023
Shahid Bahonar University of Kerman
Kerman, Iran
- Graduate workshop, Fall 2023.
Teaching Assistant — Fall 2023
Shahid Bahonar University of Kerman
Kerman, Iran
- Advanced Programming (C++) — Dr. Behnam Mohammad Hassani Zadeh.
- Fundamentals of Programming (Pascal) — Dr. Mohammad Masoud Javidi.
- Topics in Data Science (graduate course) — Dr. Najme Mansouri.
Research
Publications
Peer-reviewed papers, preprints, and work in progress.
Under review2026
Unraveling Dropout: Classifying Dropout Methods in the Context of Research Challenges Across Data Types
Amir Mohammad Sharafaddini
Computer Science Review
This review examines how dropout methods are used across data types and research settings, organizing their design choices, strengths, and trade-offs. It highlights that regularization can behave differently for images, text, and structured data, so the choice of dropout strategy should reflect the task and model. The manuscript is currently under review.
Published2026
Dual Memory Matrix Gannet Optimization with Dynamic Weight Adjustment for Robust Feature Selection
Amir Mohammad Sharafaddini, Behnam Mohammad Hasani Zade, Najme Mansouri
Cognitive Computation · Springer Nature
This study introduces an improved gannet-based feature-selection method that keeps two complementary memories of promising solutions. It adapts the scoring weights to balance predictive performance with selecting fewer, more useful features. The authors evaluate it on benchmark functions and 17 datasets, comparing its results with established optimization methods. The approach aims to search more reliably, avoid premature convergence, and produce compact feature sets.
Published2025
Attrition Mill Optimization Algorithm: A Novel Approach for Solving Engineering Optimization Problems
Amir Mohammad Sharafaddini, Behnam Mohammad Hasani Zade, Najme Mansouri
The Journal of Supercomputing · Springer Nature
Attrition Mill Optimization (AMO) is a metaheuristic inspired by the repeated interactions inside an attrition mill. Its three stages—initialization, operation, and rejuvenation—balance broad exploration with local refinement, while opposition-based learning helps preserve diversity. The study evaluates AMO on benchmark functions and engineering design problems, comparing solution quality and computational speed with other optimization algorithms.
Published2025
A Binary Chaotic Transient Search Optimization Algorithm for Enhancing Feature Selection
Amir Mohammad Sharafaddini, Najme Mansouri
Arabian Journal for Science and Engineering · Springer Nature
This paper adapts transient search optimization to the binary choices used in feature selection. Chaotic sequences guide the search across candidate feature subsets, helping the method explore more widely and avoid redundant inputs. Evaluated on 12 UCI datasets, the approach is compared with established methods using classification performance, convergence, and diversity. It offers a focused way to select compact, informative feature sets.
Published2024
Leveraging Transfer Learning for High-Accuracy Breast Cancer Classification from Histopathological Images
Amir Mohammad Sharafaddini, Najme Mansouri
6th International Conference on Electrical Engineering, Computer, Mechanics, and Artificial Intelligence
This study explores transfer learning for classifying breast cancer from histopathological images. It adapts visual features learned from broader image datasets to tissue microscopy, aiming to make better use of labeled medical images and recognize patterns associated with cancer classes. The work evaluates this approach as a practical route toward computer-assisted analysis of breast tissue.
Published2024
Enhanced Skin Cancer Classification Using Deep Learning and Gradient Boosting Techniques
Amir Mohammad Sharafaddini, Najme Mansouri
2024 14th International Conference on Computer and Knowledge Engineering (ICCKE 2024) · IEEE
This conference paper explores a skin-cancer classification pipeline that combines deep-learning image features with gradient-boosting techniques. The approach aims to use representations learned by neural networks while drawing on the practical strengths of boosted decision trees. It evaluates this combination for distinguishing skin-cancer categories and contributes to research on computer-assisted analysis of medical images.
Published2022
An Efficient Gannet Optimization Algorithm for Feature Selection Based on Sensitivity and Specificity
Amir Mohammad Sharafaddini, Najme Mansouri
Journal of Algorithms and Computation, 54(2), 49–69
This work adapts gannet-inspired search to feature selection, where a compact feature set must still help a classifier detect both positive and negative cases. It considers sensitivity and specificity when evaluating candidate subsets, connecting the optimization process to practical classification needs. The study explores how this nature-inspired search can support more balanced feature selection on benchmark data.
Work
Projects
Research projects, experiments, and open-source tools.
Pantry Meals
A macOS app that suggests meals from ingredients already in your pantry. A LangGraph agent searches a PostgreSQL collection of 501 recipes and lists the ingredients you still need for each suggestion.
Lightweight NAS CNN
A computer-vision project exploring neural architecture search for compact convolutional networks. It focuses on finding a practical balance between model size and predictive performance, so a selected network can be efficient as well as capable. The repository demonstrates a search-based approach to designing lightweight CNN architectures for image-classification tasks.
Adaptive Traffic DQN
A reinforcement-learning simulation for traffic lights that learns signal actions from changing road conditions. Using a Deep Q-Network, the controller adapts its choices as traffic shifts, with the goal of reducing waiting and congestion. The project offers a compact testbed for exploring how learned traffic control behaves across different simulated scenarios.
Evidence-Based RAG
An experimental retrieval-augmented generation pipeline that searches for relevant passages before producing an answer. It is designed to keep responses tied to supporting evidence, making it easier to inspect where an answer came from and reduce unsupported claims. The project explores similarity-based retrieval and evidence-aware response generation for research and question-answering workflows.