Amin Ahsan Ali, PhD

Abu Tyeb Azad, Fahim Ahmed, Ishita Sur Apan, Ezharuddin Jubaer, Sumaiya Karim Katha, Armun Alam, Amin Ahsan Ali, Aman Chadha, Md Mofijul Islam, AKM Mahbubur Rahman

Khondo: A Multimodal Benchmark for Document Packet Splitting of Bangla Forms

arXiv (Cornell University), 2026

• multimodal benchmark
• vision-language models
• document segmentation
• document clustering
• administrative documents

Aldrin Kabya Biswas, Md Fahim, Md Tahmid Hasan Fuad, Akm Moshiur Rahman Mazumder, Amin Ahsan Ali, AKM Mahbubur Rahman

Training-free layer selection for partial fine-tuning of language models

Information Sciences 740, pp. 123204, 2026

• partial fine-tuning
• domain generalization
• pre-trained language models
• text generation
• layer selection
• parameter-efficient fine-tuning
• sentence classification
• transformer layers

Aldrin Kabya Biswas, Md Fahim, Md Ashraful Amin, Amin Ahsan Ali, AKM Mahbubur Rahman

PRiSM: Partial Ranking via Inter-layer Semantic Measurement for Efficient Fine-tuning of Language Models

Proceedings of the Fifteenth Language Resources and Evaluation Conference (LREC 2026) 11(16), pp. 10313-10323, 2026

• partial rankings
• language model fine-tuning
• efficient fine-tuning
• Semantic similarity

Amin Masud Ali, Muntasir Wahed, Amin Ahsan Ali, Moinul Zaber

Capturing the spatiotemporal inequality in electricity consumption at the subnational level of Bangladesh using nighttime lights

Journal of the Asia Pacific Economy 31(2), pp. 397-426, 2026

• VIIRS Day/Night Band
• electricity consumption
• built-up area density
• nighttime lights
• rural electrification
• urban-rural divide
• subnational inequality
• Bangladesh

• SDG 10Reduced Inequalities
• SDG 7Affordable and Clean Energy

Md Fahim, Md Farhan Ishmam, Mir Sazzat Hossain, M Ashraful Amin, Amin Ahsan Ali, AKM Rahman

R-MMA: Enhancing Vision-Language Models with Recurrent Adapters for Few-Shot and Cross-Domain Generalization

Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 6496-6506, 2026

In: 2026 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

• vision-language models
• domain generalization
• cross-dataset generalization
• self-attention
• few-shot learning
• CLIP
• parameter efficiency
• cross-modal consistency

Abu Tyeb Azad, Ishita Sur Apan, Fahim Ahmed, Sumaiya Karim Katha, Ezharuddin Jubaer, Armun Alam, Pranjal Kumar Nandi, Amin Ahsan Ali, Aman Chadha, Md Mofijul Islam, AKM Mahbubur Rahman

BaFCo: A Document Understanding Benchmark for Complex Bangla Form Comprehension

Computer Vision – ECCV 2026, pp. 402-420, 2026

• key information extraction
• multimodal large language models
• document layout analysis
• government forms
• chain-of-thought prompting
• low-resource languages
• zero-shot prompting
• document understanding
• Bengali language
• annotated dataset

Faisal Hossain Raquib, Akm Moshiur Rahman Mazumder, Md Fahim, Md Tahmid Hasan Fuad, Md Farhan Ishmam, Faria Sultana, M Ashraful Amin, Amin Ahsan Ali, Akmmahbubur Rahman

BanHADEX: Towards Explainable HAte Speech Detection in Bangla Using Human Annotated EXplanation

Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 43652-43674, 2026

• hate speech detection
• Explainable AI
• Bengali language

• SDG 16Peace, Justice and Strong Institutions

Md Raqibul Islam, Adrita Khan, Mir Sazzat Hossain, Choudhury Ben Yamin Siddiqui, Md Zakir Hossan, Tanjib Khan, M Arshad Momen, Amin Ahsan Ali, AKM Mahbubur Rahman

KIGNet: Physics-Motivated Multi-Graph Representation Learning for Explainable Jet Tagging

arXiv (Cornell University), 2025

• jet tagging
• graph neural networks
• Lund jet plane
• explainable deep learning
• perturbative QCD factorization
• jet classification
• Grad-CAM
• kinematic parameters

Md Akil Raihan Iftee, Syed Md Ahnaf Hasan, Amin Ahsan Ali, AKM Mahbubur Rahman, Sajib Mistry, Aneesh Krishna

FedPoisonTTP: A Threat Model and Poisoning Attack for Federated Test-Time Personalization

arXiv (Cornell University), pp. 3330-3339, 2025

• data poisoning
• Federated learning
• surrogate models
• feature consistency
• Threat model
• adversarial queries

Md Akil Raihan Iftee, Mir Sazzat Hossain, Rakibul Hasan Rajib, Tariq Iqbal, Md Mofijul Islam, M Ashraful Amin, Amin Ahsan Ali, AKM Rahman

SloMo-Fast: Slow-Momentum and Fast-Adaptive Teachers for Source-Free Continual Test-Time Adaptation

arXiv (Cornell University), 2025

• continual test-time adaptation
• catastrophic forgetting
• source-free adaptation
• domain shift