Hitting your MARQ: Multimodal ARgument Quality Assessment in Long Debate Video

no code implementations • EMNLP 2021 • Md Kamrul Hasan, James Spann, Masum Hasan, Md Saiful Islam, Kurtis Haut, Rada Mihalcea, Ehsan Hoque

The combination of gestures, intonations, and textual content plays a key role in argument delivery.

CoDesc: A Large Code–Description Parallel Dataset

1 code implementation • Findings (ACL) 2021 • Masum Hasan, Tanveer Muttaqueen, Abdullah Al Ishtiaq, Kazi Sajeed Mehrab, Md. Mahim Anjum Haque, Tahmid Hasan, Wasi Ahmad, Anindya Iqbal, Rifat Shahriyar

Auto-Gait: Automatic Ataxia Risk Assessment with Computer Vision on Gait Task Videos

1 code implementation • 15 Mar 2022 • Wasifur Rahman, Masum Hasan, Md Saiful Islam, Titilayo Olubajo, Jeet Thaker, Abdelrahman Abdelkader, Phillip Yang, Tetsuo Ashizawa, Ehsan Hoque

In this paper, we investigated whether we can 1) detect participants with ataxia-specific gait characteristics (risk-prediction), and 2) assess severity of ataxia from gait (severity-assessment) using computer vision.

Clinical Knowledge Feature Importance

CoDesc: A Large Code-Description Parallel Dataset

1 code implementation • 29 May 2021 • Masum Hasan, Tanveer Muttaqueen, Abdullah Al Ishtiaq, Kazi Sajeed Mehrab, Md. Mahim Anjum Haque, Tahmid Hasan, Wasi Uddin Ahmad, Anindya Iqbal, Rifat Shahriyar

In this study, we present CoDesc -- a large parallel dataset composed of 4. 2 million Java methods and natural language descriptions.

Code Search Code Summarization +2

Text2App: A Framework for Creating Android Apps from Text Descriptions

2 code implementations • 16 Apr 2021 • Masum Hasan, Kazi Sajeed Mehrab, Wasi Uddin Ahmad, Rifat Shahriyar

We overcome this limitation by transforming natural language into an abstract intermediate formal language representing an application with a substantially smaller number of tokens.

Code Generation Language Modelling

BERT2Code: Can Pretrained Language Models be Leveraged for Code Search?

no code implementations • 16 Apr 2021 • Abdullah Al Ishtiaq, Masum Hasan, Md. Mahim Anjum Haque, Kazi Sajeed Mehrab, Tanveer Muttaqueen, Tahmid Hasan, Anindya Iqbal, Rifat Shahriyar

In this work, we leverage the efficacy of these embedding models using a simple, lightweight 2-layer neural network in the task of semantic code search.

Code Search Natural Language Queries

Not Low-Resource Anymore: Aligner Ensembling, Batch Filtering, and New Datasets for Bengali-English Machine Translation

1 code implementation • EMNLP 2020 • Tahmid Hasan, Abhik Bhattacharjee, Kazi Samin, Masum Hasan, Madhusudan Basak, M. Sohel Rahman, Rifat Shahriyar

With the segmenter and the two methods combined, we compile a high-quality Bengali-English parallel corpus comprising of 2. 75 million sentence pairs, more than 2 million of which were not available before.

Machine Translation Sentence segmentation +1