Papers are sorted by the most recent date. The full list of articles is available on my Google Scholar.
2026
1. BanCyB: A Hybrid Transformer-LSTM Framework for Explainable Multi-Label Cyberbullying Detection in Low-Resource Setting
Discover Applied Sciences (Springer, Journal)
Authors - M.M. Hasan, M.S. Arefin, Md. Ayon Mia, P.K. Dhar
Accepted
2025
2. BANMIME: Misogyny Detection with Metaphor Explanation on Bangla Memes
Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (EMNLP 2025), Main Conference [Paper]
Authors - Md. Ayon Mia, A.M.R. Mazumder, K.S. Sayma, M. Fahim, M.T.H. Fuad, A.M.M. Rahman, M.I. Khan
Published: November 2025
3. BANHATEME:Understanding Hate in Bangla Memes through Detection, Categorization, and Target Profiling
Second Bangla Language Processing Workshop (BLP 2025), co-located with ICJNLP-AACL 2025 [Preprint]
Authors - Md. Ayon Mia, M. Fahim
Accepted
4. Enhanced Semantic Relatedness Prediction Using XLM‐RoBERTa and CNNs With K‐Fold Cross‐Validation
The Journal of Engineering (JoE), Wiley & The Institution of Engineering and Technology [Paper]
Authors - Md. Ayon Mia, Fahim Shakil Tamim, Zaima Sartaj Taheri, Md. Alamin Talukder
Published: June 2025
5. Prompting vs Fine-Tuning for Bengali Violence Detection in Low-Resource Settings
28th International Conference on Computer and Information Technology (ICCIT 2025) [Preprint]
Authors - Md. Ayon Mia, A.Hasan, A. Rahman, G.S.M. Mursalin, M.I. Khan
Accepted
6. SETBoost: An Interpretable Machine Learning Approach for Predicting Software Employee Turnover Tendency
28th International Conference on Computer and Information Technology (ICCIT 2025) [Preprint]
Authors - A. Rahman, M.I. Hossain, Md. Ayon Mia, M.I. Khan
Accepted
7. Vision Transformers for Multi-Class Eye Disease Classification: Enhancing Early Detection in Resource-Constrained Healthcare Settings
3rd International Conference on Big Data, IoT and Machine Learning (BIM 2025) [Preprint]
Authors - A. Rahman, M. Hasan, Md. Ayon Mia, M.I. Khan
Accepted
8. Fraud Detection with DistilBERT: A Transformer-Based Approach to Behavioral Banking Sequences
3rd International Conference on Big Data, IoT and Machine Learning (BIM 2025) [Preprint]
Authors - M.M. Hasan, A.A. Mahin, S. Chakraborty, M. Afrose, Md. Ayon Mia, M.A. Based
Accepted: 2025
9. Explainable Self-Attentive Transformer Model for Bangla Mental Health Disorder Detection
International Conference on Intelligent Data Analysis and Applications (IDAA 2025) [Preprint]
Authors - A.S.H. Hridoy, N.S. Shohan, M.S. Alif, N.M.A. Mahin, Md. Ayon Mia
Accepted: 2025
10. KCRL@DravidianLangTech 2025: Multi-View Feature Fusion with XLM-R for Tamil Political Sentiment Analysis
DravidianLangTech Workshop @ NAACL 2025 [Paper]
Authors - Md. Ayon Mia, Fariha Haq, Md. Tanvir Ahammed Shawon, Golam Sarwar Md. Mursalin, Muhammad Ibrahim Khan
11. KCRL@DravidianLangTech 2025: Multi-Pooling Feature Fusion with XLM-RoBERTa for Malayalam Fake News Detection and Classification
DravidianLangTech Workshop @ NAACL 2025 [Paper]
Authors - Fariha Haq, Md. Tanvir Ahammed Shawon, Md. Ayon Mia, Golam Sarwar Md. Mursalin, Muhammad Ibrahim Khan
12. ID4Fusion@CASE 2025: A Multimodal Approach to Hate Speech Detection in Text-Embedded Memes Using Ensemble Transformer-Based Approach
The 8th Workshop on Challenges and Applications of Automated Extraction of Socio-Political Events from Text [Preprint]
Authors - T.B. Rashfi, M.T.A. Shawon, Md. Ayon Mia, M.I. Khan
Accepted: 2025
13. PhantomTroupe@CASE 2025: Multimodal Hate Speech Detection in Text-Embedded Memes using Instruction-Tuned LLMs
The 8th Workshop on Challenges and Applications of Automated Extraction of Socio-Political Events from Text [Preprint]
Authors - Amin, M.A. Horaira, M.T.A. Shawon, Md. Ayon Mia, M.I. Khan
Accepted: 2025
14. CUET_KCRLatCheckThat!2025: EnsembleNet with RoBERTa-Large for Subjectivity Detection in News Articles
CLEF 2025 Conference and Labs of the Evaluation Forum [Paper]
Authors - M.T.A. Shawon, F. Haq, Md. Ayon Mia, G.S.M. Mursalin, M.I. Khan
2024
15. Golden_Duck at #SMM4H 2024: A Transformer-based Approach to Social Media Text Classification
ACL 2024 - SMM4H Workshop [Paper]
Authors - Md. Ayon Mia, M. Yahan, H. Murad, M.I. Khan
2023
16. EmptyMind at BLP-2023 Task 1: A Transformer-based Hierarchical-BERT Model for Bangla Violence-Inciting Text Detection
BLP Workshop @ EMNLP 2023 [Best Paper Award] [Paper]
Authors - U Das, K Fatema, Md. Ayon Mia, M Yahan, MS Mowla, MDF Ullah, A Sarker, H Murad
17. EmptyMind at BLP-2023 Task 2: Sentiment Analysis of Bangla Social Media Posts using Transformer-Based Models
BLP Workshop @ EMNLP 2023 [Paper]
Authors - K Fatema, U Das, Md. Ayon Mia, MS Mowla, M Yahan, M.F. Ullah, A Sarker, H Murad
Under Review Papers
18. Code-Mixed Language Translation: An Integrated Framework with Enhanced Preprocessing and N-gram Optimization for Cross-Lingual Communication
Engineering Applications of Artificial Intelligence (Elsevier, Journal)
Authors - Md. Ayon Mia, M. Fahim, G.S.M. Mursalin, M.J.I. Basher, H. Murad, M.R. Islam, M.I. Khan, I.H. Sarker
19. BanGRev: A novel Bengali tech gadget review corpus utilizing triple-pooling feature fusion with explainable AI
New Generation Computing (Springer, Journal)
Authors - Md. Ayon Mia, A. Rahman, G.S.M. Mursalin, M.I. Khan, I.H. Sarker
20. Addressing Rating-Review Discrepancy in a Novel Dataset: An Explainable Pharmaceutical Product Evaluation using Multi-Stream Attention Transformer
Neural Computing and Applications (Springer, Journal)
Authors - A. Rahman, Md. Ayon Mia, M.I. Khan, I.H. Sarker
Research by Topic
Natural Language Processing for Low-Resource Languages
- Code-mixed language translation for low-resource languages
- Back-transliteration
- Social media text classification
- Violence-inciting text detection in Bangla
- Fake news detection in Dravidian languages
- Sentiment analysis for Bangla social media posts
- Semantic relatedness prediction in English language
Transformer-based Models & Techniques
- Multilingual LMs (XLM-RoBERTa, mBERT, MuRIL, IndicBERT) for cross-lingual applications
- Feature fusion techniques for improved performance
- Multi-view and multi-pooling strategies
- BERT adaptations for low-resource languages
- Hierarchical transformer architectures
