{firstName}kh@amazon.com {firstName}{lastInital}@berkeley.edu About | Publications | Talks | Teaching | Google Scholar

About

I am a senior applied scientist at Amazon, working on the AGI team.

Previously, I was a senior researcher at Microsoft, working on the Microsoft Translator team. Before that, I was PhD student in Computer Science at The Johns Hopkins University (JHU), advised by Philipp Koehn. I was part of the Center for Language and Speech Processing (CLSP) and the machine translation group.

In summer 2019, I was a research intern at Lilt, working on translator-in-the-loop machine translation.

I graduated from UC Berkeley with a B.A. in Computer Science in May 2015. I worked with Colleen Lewis on computer science education, Andreas Stolcke on tools for speech processing, and Jerry Feldman on natural language understanding.

Education

The Johns Hopkins University | August 2015 - 2021

• PhD, Computer Science

• MSE, Computer Science | May 2017

University of California, Berkeley | August 2011 - May 2015

• BA, Computer Science - Applied Language Studies Minor

Publications

* indicates authors contributed equally.
Please see Google Scholar for latest publications.

• Adapters for Altering LLM Vocabularies: What Languages Benefit the Most?
  HyoJung Han, Akiko Eriguchi, Haoran Xu, Hieu Hoang, Marine Carpuat, Huda Khayrallah
  To appear at The International Conference on Learning Representations (ICLR) 2025
  [preprint]

• X-ALMA: Plug & Play Modules and Adaptive Rejection for Quality Translation at Scale [ICLR Spotlight – top 5.1% of submitted papers.]
  Haoran Xu, Kenton Murray, Philipp Koehn, Hieu Hoang, Akiko Eriguchi, Huda Khayrallah
  To appear at The International Conference on Learning Representations (ICLR) 2025
  [preprint]

• On-the-Fly Fusion of Large Language Models and Machine Translation
  Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL) 2024
  Hieu Hoang, Huda Khayrallah, and Marcin Junczys-Dowmunt.
  [paper]

• Improving Statistical Significance in Human Evaluation of Automatic Metrics via Soft Pairwise Accuracy
  Brian Thompson, Nitika Mathur, Daniel Deutsch, and Huda Khayrallah
  Proceedings of the Ninth Conference on Machine Translation (WMT) 2024
  [paper]

• Perplexity-Driven Case Encoding Needs Augmentation for CAPITALIZATION Robustness
  Rohit Jain, Huda Khayrallah, Roman Grundkiewicz, Marcin Junczys-Dowmunt
  Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (IJCAI-AACL) 2023
  [paper] [slides.pdf]

• Findings of the WMT 2023 Shared Task on Parallel Data Curation
  Steve Sloto, Brian Thompson, Huda Khayrallah, Tobias Domhan, Thamme Gowda, Philipp Koehn
  Proceedings of the Eighth Conference on Machine Translation (WMT) 2023
  [paper]

• Revisiting Locality Sensitive Hashing for Vocabulary Selection in Fast Neural Machine Translation
  Hieu Hoang, Marcin Junczys-dowmunt, Roman Grundkiewicz, Huda Khayrallah
  Proceedings of the Seventh Conference on Machine Translation (WMT) 2023
  [paper]

• SOTASTREAM: A Streaming Approach to Machine Translation Training
  Proceedings of the 3rd Workshop for Natural Language Processing Open Source Software (NLP-OSS) 2023
  Matt Post, Thamme Gowda, Roman Grundkiewicz, Huda Khayrallah, Rohit Jain, Marcin Junczys-Dowmunt
  [paper]

• Doubly-Trained Adversarial Data Augmentation for Neural Machine Translation
  Weiting Tan, Shuoyang Ding, Huda Khayrallah, Philipp Koehn
  Proceedings of the 15th biennial conference of the Association for Machine Translation in the Americas (AMTA) 2022
  [paper]

• Measuring the ‘I don’t know’ Problem through the Lens of Gricean Quantity
  Huda Khayrallah, João Sedoc
  Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL) 2021
  [paper] [video]

• The JHU Submission to the 2020 Duolingo Shared Task on Simultaneous Translation and Paraphrase for Language Education
  Huda Khayrallah, Jacob Bremerman, Arya D. McCarthy, Kenton Murray, Winston Wu, Matt Post
  Proceedings of the Fourth Workshop on Neural Generation and Translation at ACL 2020
  [paper] [video]

• Simulated Multiple Reference Training Improves Low-Resource Machine Translation [Nominated for Best Paper at WeCNLP]
  Huda Khayrallah, Brian Thompson, Matt Post, Philipp Koehn
  Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) 2020
  [paper] [code] [data] [EMNLP_slides.pdf] [EMNLP_slides.ppt]
  [3 min WeCNLP video poster] [8 min WeCNLP lightning talk video] [wecnlp_slides.pdf] [WeCNLP_slides.ppt] [wecnlp_poster.pdf] [WeCNLP_poster.pptx]

• SMRT Chatbots: Improving Non-Task-Oriented Dialog with Simulated Multiple Reference Training
  Huda Khayrallah, João Sedoc
  Findings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP) 2020
  [paper]

• The JHU Submission to the 2020 Duolingo Shared Task on Simultaneous Translation and Paraphrase for Language Education
  [Highest scoring submission in all 5 language pairs]
  Huda Khayrallah, Jacob Bremerman, Arya D. McCarthy, Kenton Murray, Winston Wu, Matt Post
  Proceedings of the Fourth Workshop on Neural Generation and Translation at ACL 2020
  [paper] [bib] [slides.ppt] [slides.pdf]

• HABLex: Human Annotated Bilingual Lexicons for Experiments in Machine Translation
  Brian Thompson,* Rebecca Knowles,* Xuan Zhang,* Huda Khayrallah, Kevin Duh, Philipp Koehn
  Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) 2019
  [paper] [bib] [data]

• Deep Generalized Canonical Correlation Analysis
  Adrian Benton, Huda Khayrallah, Biman Gujral, Dee Ann Reisinger, Sheng Zhang, Raman Arora
  Proceedings of the 4th Workshop on Representation Learning for NLP (RepL4NLP) 2019 at ACL
  [paper] [bib] [code]

• Overcoming Catastrophic Forgetting During Domain Adaptation of Neural Machine Translation
  Brian Thompson, Jeremy Gwinnup, Huda Khayrallah, Kevin Duh, Philipp Koehn
  Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL) 2019
  [paper] [bib]

• Improved Lexically-Constrained Decoding for Translation and Monolingual Rewriting
  J. Edward Hu, Huda Khayrallah, Ryan Culkin, Patrick Xia, Tongfei Chen, Matt Post, Benjamin Van Durme
  Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL) 2019
  [paper] [bib] [appendix.pdf] [demo & code]

• An Interactive Teaching Tool for Introducing Novices to Machine Translation
  Huda Khayrallah, Rebecca Knowles, Kevin Duh, Matt Post
  Proceedings of the 50th ACM Technical Symposium on Computer Science Education (SIGCSE) 2019
  [paper] [bib] [poster.pdf] [demo & code]

• Freezing Subnetworks to Analyze Domain Adaptation in Neural Machine Translation
  Brian Thompson, Huda Khayrallah, Antonios Anastasopoulos, Arya D. McCarthy, Kevin Duh, Rebecca Marvin, Paul McNamee, Jeremy Gwinnup, Tim Anderson, Philipp Koehn
  Proceedings of the Third Conference on Machine Translation (WMT) 2018 at EMNLP
  [paper] [bib] [slides.pdf]

• Findings of the WMT 2018 Shared Task on Parallel Corpus Filtering
  Philipp Koehn, Huda Khayrallah, Kenneth Heafield, Mikel L. Forcada
  Proceedings of the Third Conference on Machine Translation (WMT) 2018 at EMNLP
  [paper] [bib]

• The JHU Parallel Corpus Filtering Systems for WMT 2018
  Huda Khayrallah, Hainan Xu, Philipp Koehn
  Proceedings of the Third Conference on Machine Translation (WMT) 2018 at EMNLP
  [paper] [bib]

• On the Impact of Various Types of Noise on Neural Machine Translation [Outstanding Contribution Award]
  Huda Khayrallah, Philipp Koehn
  Proceedings of the Workshop on Neural Machine Translation (WNMT) 2018 at ACL
  [paper] [bib] [slides.ppt] [slides.pdf] [poster.pdf] [data]

• Regularized Training Objective for Continued Training for Domain Adaptation in Neural Machine Translation
  Huda Khayrallah, Brian Thompson, Kevin Duh, Philipp Koehn
  Proceedings of the Workshop on Neural Machine Translation (WNMT) 2018 at ACL
  [paper] [bib] [poster.pdf]

• Improving Low Resource Machine Translation using Morphological Glosses
  Steven Shearing, Christo Kirov, Huda Khayrallah, David Yarowsky
  Proceedings of the Conference of the Association for Machine Translation in the Americas (AMTA) 2018
  [paper] [bib]

• Neural Lattice Search for Domain Adaptation in Machine Translation
  Huda Khayrallah, Gaurav Kumar, Kevin Duh, Matt Post, Philipp Koehn
  Proceedings of the International Joint Conference on Natural Language Processing (IJCNLP) 2017
  [paper] [bib] [code] [slides.ppt] [slides.pdf]

• The JHU Machine Translation Systems for WMT 2017
  Shuoyang Ding, Huda Khayrallah, Philipp Koehn, Gaurav Kumar, Matt Post, Kevin Duh
  Proceedings of the Second Conference on Machine Translation (WMT) 2017
  [paper] [poster] [bib]

• Paradigm Completion for Derivational Morphology
  Ryan Cotterell, Ekaterina Vylomova, Huda Khayrallah, Christo Kirov, David Yarowsky
  Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP) 2016
  [paper] [bib] [data]

• Translation of Unknown Words in Low Resource Languages
  Biman Gujral, Huda Khayrallah, Philipp Koehn
  Proceedings of the Conference of the Association for Machine Translation in the Americas (AMTA) 2016
  [paper] [bib] [slides.ppt] [slides.pdf]

• The JHU Machine Translation Systems for WMT 2016
  Shuoyang Ding, Kevin Duh, Huda Khayrallah, Philipp Koehn, Matt Post
  Proceedings of the First Conference on Machine Translation (WMT) 2016
  [paper] [poster] [bib]

• Natural Language For Human Robot Interaction
  Huda Khayrallah, Sean Trott, Jerome Feldman
  Workshop on Human-Robot Teaming at the ACM/IEEE conference on Human-Robot Interaction (HRI) 2015
  [paper] [bib]

• Mining Data from the AP CS A Exam: Patterns, Non-Patterns, and Replication Failure
  Colleen Lewis, Huda Khayrallah, Amy Tsai
  International Computing Education Research Conference (ICER) 2013
  [paper] [bib]

Unrefereed Reports

• An Empirical Exploration of Curriculum Learning for Neural Machine Translation
  Xuan Zhang, Gaurav Kumar, Huda Khayrallah, Kenton Murray, Jeremy Gwinnup, Marianna J Martindale, Paul McNamee, Kevin Duh, and Marine Carpuat
  [paper] [bib] [code]

Unrefereed Reports

• How to Choose How to Choose Your Chatbot: A Massively Multi-System MultiReference Data Set for Dialog Metric Evaluation
  Huda Khayrallah, Zuhaib Akhtar, Edward Cohen, Jyothir S V, João Sedoc [paper]

Talks

• Machine Translation with Diverse Data Sources
  NYU Abu Dhabi Computer Science Seminar
  September 1, 2019
  [slides.ppt] [slides.pdf]

• Machine Translation with Diverse Data Sources
  University of Pennsylvania Computational Linguistics Seminar
  April 8, 2019
  [slides.ppt] [slides.pdf]

• Machine Translation with Diverse Data Sources
  Johns Hopkins University Center for Language and Speech Processing Seminar
  March 29, 2019
  [slides.ppt] [slides.pdf]

• Continued Training Algorithms
  SCALE 2018 Workshop at Johns Hopkins University
  August 9, 2018
  [slides.ppt] [slides.pdf]

• Neural Lattice Search for Domain Adaptation in Machine Translation (+ An overview of Machine Translation)
  Johns Hopkins University Center for Language and Speech Processing Seminar
  November 17, 2017
  [slides.ppt] [slides.pdf]

Teaching

• Teaching Assistant for Machine Translation (JHU CS 601.468/668) taught by Philipp Koehn – Fall 2018
  

• Teaching Assistant for Artificial Intelligence (JHU CS 601.464/664) taught by Benjamin Van Durme – Spring 2017
  

• Guest Lecturer for Artificial Intelligence (JHU CS 601.464/664) taught by Benjamin Van Durme – Fall 2018, Fall 2019
  

Huda rymes with CUDA (parallel framework) and gouda (cheese) – WHO-duh – /ˈhuːdə/
Khayrallah is less scary than it looks! kay-RA-la – /keɪˈɹɑlɑ/
My pronouns are she/her.