Publications

Peer-reviewed journal articles, conference papers, manuscripts under review, and works in preparation. Google Scholar profile →

2
Q1 Journal Articles
2
Conference Papers
2
Under Review
3
In Preparation

Peer-Reviewed Journal Articles

Method diagram for J.1 — Multi-Teacher Knowledge Distillation for Brain Tumor Classification
J.1

Multi-Teacher Knowledge Distillation and Ensemble Algorithm for Efficient Brain Tumor Classification in Resource-Constrained Environments with Explainable AI

Md. Samiul Alim, Tamim, M. S., Sarkar, S., Yousuf, M. A., Al-Moisheer, A. S., Alyami, S. A., & Moni, M. A.
Knowledge-Based Systems, 2025 Q1 IF 7.6
DOI
Method diagram for J.2 — Rainfall Classification in Bangladesh
J.2

Ensemble and Temporal Feature-Based Framework for Rainfall Classification in Bangladesh

Tamim, M. S., Md. Samiul Alim, Khan, T. A., Rahman, M., & Anwar, M. M.
PLOS ONE, 2026 Q1 IF 2.6
Article

Peer-Reviewed Conference Papers

Method diagram for C.1 — Teacher-Guided One-Shot Pruning
C.1

Teacher-Guided One-Shot Pruning via Context-Aware Knowledge Distillation

Md. Samiul Alim, Khan, S., Biswas, A., Rahman, F., Rahman, S., & Mohammed, N.
IEEE BigData 2025, Macau, China CORE B 18.8% acceptance
IEEE Xplore
Method diagram for C.2 — Bangladesh 2024 Mass Uprising sentiment analysis
C.2

When a Nation Speaks: Machine Learning and NLP in People's Sentiment Analysis during Bangladesh's 2024 Mass Uprising

Md. Samiul Alim, Mahir, S. T., Maisha, R., Tanvir, A. K., & Md Mushfique, A.
IEEE International Conference on Computer and Information Technology (ICCIT 2025)
arXiv

Manuscripts Under Review

Method diagram for U.1 — CAT-GS multimodal gating
U.1

CAT-GS: Calibrated Adaptive Gating for Balanced and Robust Multimodal Learning

Under review at Neurocomputing Q1 Under Review
Method diagram for U.2 — UnRestSent200K benchmark
U.2

UnRestSent200K: Benchmarking Contextual Sentiment Understanding During Political Unrest

Submitted to KDD 2026 Under Review

Manuscripts in Preparation

Method diagram for P.1 — CLIPByte tokenizer-free captioning
P.1

CLIPByte: Open-Vocabulary Multilingual Image Captioning with Byte-Level Decoding

Targeting BMVC 2026 In Preparation

A tokenizer-free vision–language model pairing CLIP ViT-B/32 with a ByT5-small byte-level decoder for captioning across English, Thai, Japanese, Italian, German, and Vietnamese.

Method diagram for P.2 — DINO memory retrieval for TTA
P.2

Beyond CLIP: Leveraging DINO for Reliable Memory Retrieval in Cache-Based Test-Time Adaptation

Targeting BMVC 2026 In Preparation
Method diagram for P.3 — Neuroevolving Kolmogorov-Arnold Networks
P.3

Neuroevolving Kolmogorov–Arnold Networks

In preparation In Preparation