Mohamed Salem Elaraby (Salem)

mse30[At]pitt[Dot]edu   [CV]   [LinkedIn]   [GitHub]   [Scholar]

Summer 2023: I am currently working as a research intern at Tiktok Inc. in San Jose.

I am currently a second year Ph.D. student in Pitt EducaTional And Language technology lab (PETAL), The University of Pittsburgh, advised by Prof. Diane J. Litman.

My research is focused on Natural Language Processing with focus on text summarization.

From 2020 to 2021 I worked as an applied Datascientist in the CSALT team (Cairo Speech and Language Technology) at Microsoft ATLC in developing Arabic speech recognition models.

From 2018 to 2020 I worked as a Datascientist II at Raisa Energy LLC. working on various datascience tasks

I received my Bachelors (Bsc) in 2014 from Cairo University in Communications and Computer Engineering, and my Master of science (Msc) in 2017 in Electronics and Computer Communications from the same University.


  • Mohamed Elaraby,Yang Zhong and Diane Litman "Towards Argument-Aware Abstractive Summarization of Long Legal Opinions with Summary Reranking" Fingings ACL.2023[pdf]
  • Mohamed Elaraby, and Diane Litman "ArgLegalSumm: Improving Abstractive Summarization of Legal Documents with Argument Mining." COLING .2022[pdf]
  • Mohamed Elaraby, and Diane Litman "Self-trained Pretrained Language Models for Evidence Detection" ArgMining @EMNLP .2021[pdf]
  • Ahmed Magooda,Mohamed Elaraby, and Diane Litman "Exploring Multitask Learning for Low-Resource AbstractiveSummarization" Findings @ EMNLP 2021 [pdf]
  • Mohamed Elaraby, and Ahmed Ismail Zahran "A Character Level Convolutional BiLSTM for Arabic Dialect Identification" WANLP @ ACL 2019 [pdf]
  • Mohamed Elaraby, and Muhammad-Abdulmageed "Deep Models for Arabic Dialect Identification on Benchmarked Data" Vardial @ COLING 2018 [pdf]
  • Hassan Alhuzali , Mohamed Elaraby, and Muhammad-Abdulmageed "UBC-NLP at IEST 2018: Learning Implicit Emotion With an Ensemble of Language Models" WASSA @ EMNLP 2018 [pdf]
  • Muhammad-AbdulMageed, Hassan Alhuzali , and Mohamed Elaraby, "You Tweet What You Speak: A City-Level Dataset of Arabic Dialects" LREC 2018 [pdf]
  • Hany Ahmed , Mohamed Elaraby, Abdullah M. Moussa, Mustafa Abdullah, Sherif M. Abdou, and Mohsen Rashwan "An Unsupervised Speaker Clustering Technique based on SOM and I-vectors for Speech Recognition Systems" WANLP @EACL 2017 [pdf]
  • Mohamed Elaraby, Mustafa Abdullah, Sherif M. Abdou and Mohsen Rashwan "A Deep Neural Networks (DNN) Based Models for a Computer Aided Pronunciation Learning System" SPECOM 2016 [Link]

Selected Research Projects


CASUM aims at summarizing case decisions in terms of legal argument triples: the major issues a court addressed in the case, the court's conclusion with respect to each issue, and the court's reasons for reaching the conclusion. Given the legal case, we extract the legal triplets which aid the abstractive summarization process to include the legal arguments described in the document.

Course Mirror

CourseMIRROR facilitates learning at scale by enabling students to reflect in-situ via mobile devices. CourseMIRROR also helps instructors to capture the gist of reflections effectively via natural language processing.