Ahsanullah University of Science and Technology (AUST)
I am a CS graduate working in Natural Language Processing (NLP), with a focus on Bengali and other
low-resource language technologies. My research spans Bengali mathematical reasoning, regional dialect
processing, named entity recognition, and harmful content detection. More recently, I have developed a
growing interest in LLM reasoning and AI safety, particularly in understanding how large
language models reason, fail, and behave in safety-critical settings.
🎓 I am actively looking for PhD opportunities for Fall 2027 in NLP, LLM reasoning, and AI safety.
My long-term goal is to contribute to building robust, interpretable, and reliable AI systems, especially
for low-resource and multilingual settings. If my research interests align with your group, I would be glad
to connect 📩.
Warning
Problem: The current name of your GitHub Pages repository ("Solution: Please consider renaming the repository to "
http://".
However, if the current repository name is intended, you can ignore this message by removing "{% include widgets/debug_repo_name.html %}" in index.html.
Action required
Problem: The current root path of this site is "baseurl ("_config.yml.
Solution: Please set the
baseurl in _config.yml to "Education
-
Ahsanullah University of Science and Technology (AUST)B.Sc. in Computer Science and Engineering (CSE)Dec. 2020 - Jan. 2025
Experience
-
Southeast University, Dhaka, BangladeshAdjunct Lecturer23rd February, 2025 – Present -
ELITE Research Lab LLC, RemoteResearch Assistant25th May, 2026 – Present
Honors & Awards
-
Received a research grant from the Institute of Research & Training (IRT), Southeast University for ANCHOLIK-NER.2025
-
Dean's List of Honor, 2nd Position2024
-
Consistently received merit based waivers in the 1st, 2nd, 3rd, 5th, 6th, and 7th semesters based on top academic ranking and merit within the CSE Department.2020 - 2024
-
Scholars' School and College Science Fair 2nd Runners up2016
-
General Scholarship in JSC (Junior School Certificate)2014
News
Selected Publications (view all )
ANCHOLIK-NER: A benchmark dataset for bangla regional named entity recognition
Bidyarthi Paul; Faika Fairuj Preotee; Shuvashis Sarker; Shamim Rahim Refat; Shifat Islam; Tashreef Muhammad; Mohammad Ashraful Hoque; Shahriar Manzoor.
PLOS ONE 2026
TL;DR: We introduce ANCHOLIK-NER, the first benchmark dataset for Bangla regional named entity recognition, and show that BanglaBERT provides the strongest baseline across five major dialect regions.
#Bangla NLP#Named Entity Recognition#Regional Dialects#Dataset#BanglaBERT
BIDWESH: A Bangla regional based hate speech detection dataset
Azizul Hakim Fayaz; MD Uddin; Rayhan Uddin Bhuiyan; Zakia Sultana; Md Samiul Islam; Bidyarthi Paul; Tashreef Muhammad; Shahriar Manzoor.
arXiv 2025
TL;DR: We introduce BIDWESH, the first multi-dialectal Bangla hate speech dataset, covering Barishal, Noakhali, and Chittagong dialects to support dialect-aware hate speech detection and fairer content moderation.
#Bangla NLP#Hate Speech Detection#Regional Dialects#Dataset#Low-resource NLP
Leveraging large language models for bengali math word problem solving with chain of thought reasoning
Bidyarthi Paul; Jalisha Jashim Era; Mirazur Rahman Zim; Tahmid Sattar Aothoi; Faisal Muhammad Shah.
ACM TALLIP 2025
TL;DR: We introduce SOMADHAN, a Bengali math word problem dataset with step-by-step solutions, and show that chain-of-thought prompting substantially improves LLM performance on complex Bengali mathematical reasoning tasks.
#Bangla NLP#Math Word Problems#Large Language Models#Chain-of-Thought#Reasoning
Empowering bengali education with ai: Solving bengali math word problems through transformer models
Jalisha Jashim Era; Bidyarthi Paul; Tahmid Sattar Aothoi; Mirazur Rahman Zim; Faisal Muhammad Shah.
ICCIT 2024
TL;DR: We solve Bengali math word problems using transformer-based models and introduce the PatiGonit dataset, showing that mT5 achieves the best performance for translating problems into equations.
#Bangla NLP#Math Word Problems#Transformer Models#Bengali Education#Low-resource NLP
Bridging dialects: Translating standard bangla to regional variants using neural models
Md Arafat Alam Khandaker; Ziyan Shirin Raha; Bidyarthi Paul; Tashreef Muhammad.
ICCIT 2024
TL;DR: We translate standard Bangla into multiple regional dialects using neural machine translation models and show that BanglaT5 performs best for capturing dialect-specific linguistic variations.
#Bangla NLP#Dialect Translation#Neural Machine Translation#BanglaT5#Low-resource NLP
Analyzing emotions in Bangla social media comments using machine learning and LIME
Bidyarthi Paul; SM Musfiqur Rahman; Dipta Biswas; Md Ziaul Hasan; Md Zahid Hossain.
MIET 2024
TL;DR: We analyze Bangla social media comments for emotion detection using machine learning and deep learning models, and use LIME to explain model predictions for better interpretability.
#Bangla NLP#Emotion Analysis#Machine Learning#LIME#Social Media Analysis
Improving Bangla regional dialect detection using BERT, LLMs, and XAI
Bidyarthi Paul; Faika Fairuj Preotee; Shuvashis Sarker; Tashreef Muhammad.
COMPAS 2024
TL;DR: We improve Bangla regional dialect detection by comparing BERT-based models and large language models, and use LIME to interpret model predictions across multiple dialect regions.
#Bangla NLP#Dialect Detection#BERT#LLM#XAI