Raian Rahman

Dhaka, Bangladesh

Mail: raianrahman@iut-dhaka.edu

I am Raian Rahman, currently working as a Software Engineer at Therap BD Ltd, where I focus on full-stack development, system optimization, and data integration. I completed my Bachelor of Science in Computer Science and Engineering from the Islamic University of Technology(IUT) , where my research experience have primarily revolved around Natural Language Processing (NLP), Computer Vision, and Data Science.

During my time at Therap BD Ltd, I have played a key role in developing the Therap Connect application, working on Reusable UI Components ,Backend Services, RESTful APIs, and integrating IoT devices to enhance the application’s scalability and robustness. My contributions have also included optimizing database queries, which resulted in a significant reduction in execution time and improved system performance. Previously, as a Machine Learning Engineer Intern at Pioneer Alpha Limited, I worked on fine-tuning state-of-the-art NLP models like mT5 and mBART for Bengali question-answering systems.

My research work has focused on areas such as Chart Summarization, Object Detection, Multimodal Learning, Large Language Models, and Text Style Transfer with notable contributions to NLP and Computer Vision. You can find more about my research and its impact on my Google Scholar profile.

news

Sep 20, 2024	One paper accepted in EMNLP 2024!
Apr 04, 2023	One paper accepted in proceedings of Canadian AI (CANAI) 2023
Apr 01, 2023	Promoted as a Software Engineer at Therap Services

Selected Publications

Are Large Vision Language Models up to the Challenge of Chart Comprehension and Reasoning? An Extensive Investigation into the Capabilities and Limitations of LVLMs

Mohammed Saidul Islam, Raian Rahman, Ahmed Masry, Md Tahmid Rahman Laskar, Mir Tafseer Nayeem, and Enamul Hoque

arXiv preprint arXiv:2406.00257, 2024

Accepted in the Proceedings of EMNLP 2024 (Findings)

Abs Bib

Natural language is a powerful complementary modality of communication for data visualizations, such as bar and line charts. To facilitate chart-based reasoning using natural language, various downstream tasks have been introduced recently such as chart question answering, chart summarization, and fact-checking with charts. These tasks pose a unique challenge, demanding both vision-language reasoning and a nuanced understanding of chart data tables, visual encodings, and natural language prompts. Despite the recent success of Large Language Models (LLMs) across diverse NLP tasks, their abilities and limitations in the realm of data visualization remain under-explored, possibly due to their lack of multi-modal capabilities. To bridge the gap, this paper presents the first comprehensive evaluation of the recently developed large vision language models (LVLMs) for chart understanding and reasoning tasks. Our evaluation includes a comprehensive assessment of LVLMs, including GPT-4V and Gemini, across four major chart reasoning tasks. Furthermore, we perform a qualitative evaluation of LVLMs’ performance on a diverse range of charts, aiming to provide a thorough analysis of their strengths and weaknesses. Our findings reveal that LVLMs demonstrate impressive abilities in generating fluent texts covering high-level data insights while also encountering common problems like hallucinations, factual errors, and data bias. We highlight the key strengths and limitations of chart comprehension tasks, offering insights for future research.
@article{islam2024large, title = {Are Large Vision Language Models up to the Challenge of Chart Comprehension and Reasoning? An Extensive Investigation into the Capabilities and Limitations of LVLMs}, author = {Islam, Mohammed Saidul and Rahman, Raian and Masry, Ahmed and Laskar, Md Tahmid Rahman and Nayeem, Mir Tafseer and Hoque, Enamul}, journal = {arXiv preprint arXiv:2406.00257}, year = {2024}, note = {Accepted in the Proceedings of EMNLP 2024 (Findings)}, }
ChartSumm: A Comprehensive Benchmark for Automatic Chart Summarization of Long and Short Summaries

Raian Rahman, Rizvi Hasan, Abdullah Al Farhad, Md. Tahmid Rahman Laskar, Md. Hamjajul Ashmafee, and Abu Raihan Mostofa Kamal

Proceedings of the Canadian Conference on Artificial Intelligence, Jun 2023

https://caiac.pubpub.org/pub/ujhjycsw

Abs Bib

Automatic chart to text summarization is an effective tool for the visually impaired people along with providing precise insights of tabular data in natural language to the user. A large and well-structured dataset is always a key part for data driven models. In this paper, we propose ChartSumm: a large-scale benchmark dataset consisting of a total of 84,363 charts along with their metadata and descriptions covering a wide range of topics and chart types to generate short and long summaries. Extensive experiments with strong baseline models show that even though these models generate fluent and informative summaries by achieving decent scores in various automatic evaluation metrics, they often face issues like suffering from hallucination, missing out important data points, in addition to incorrect explanation of complex trends in the charts. We also investigated the potential of expanding ChartSumm to other languages using automated translation tools. These make our dataset a challenging benchmark for future research.
@article{Rahman2023ChartSumm, author = {Rahman, Raian and Hasan, Rizvi and Farhad, Abdullah Al and Laskar, Md. Tahmid Rahman and Ashmafee, Md. Hamjajul and Kamal, Abu Raihan Mostofa}, journal = {Proceedings of the Canadian Conference on Artificial Intelligence}, year = {2023}, month = jun, note = {https://caiac.pubpub.org/pub/ujhjycsw}, publisher = {Canadian Artificial Intelligence Association (CAIAC)}, title = {ChartSumm: A {Comprehensive} {Benchmark} for {Automatic} {Chart} {Summarization} of {Long} and {Short} {Summaries}}, }