PeopleofSuperintelligence

Researcher

Maja Popović

Maja Popović is a natural language processing researcher whose work connects machine translation, evaluation, and gender bias. She studied electrical engineering at the University of Belgrade, earned a PhD at RWTH Aachen, and coauthored the RWTH Machine Translation System for WMT 2009. Her work introduced chrF, set out a workflow for evaluating and improving machine translation systems, and described neural translation systems for South Slavic languages. She coauthored a method for gender-balanced conversational data and a test suite for first-person translation bias. She is a lecturer at IU International University of Applied Sciences and served on the 2026 Gender-Inclusive Translation Technologies program committee.

Now at
Dublin City University and IU International University of Applied Sciences
Works
123
Citations
3,020
h-index
22
Badges

Publication numbers from OpenAlex, as of 22 October 2026.

Timeline

  1. 2009

    The RWTH WMT system

    She graduates from the University of Belgrade’s Faculty of Electrical Engineering and earns a PhD at RWTH Aachen with a thesis on statistical machine translation and additional linguistic knowledge. In 2009, she coauthors the RWTH Machine Translation System for WMT with David Vilar, Daniel Stein, Evgeny Matusov, and Hermann Ney.

  2. 2011

    Automatic MT error analysis

    In 2011, she and Hermann Ney publish “Towards Automatic Error Analysis of Machine Translation Output” in Computational Linguistics.

  3. 2015

    The chrF metric

    In 2015, she publishes chrF, a character n-gram F-score for automatic machine-translation evaluation.

  4. 2016

    Tools for MT development

    In 2016, she coauthors a complete workflow for evaluating, comparing, and improving machine-translation systems. The toolkit combines automatic metrics and n-gram feedback with manual evaluation, error annotation, and task-based feedback.

  5. 2020

    South Slavic neural translation

    In 2020, she and Alberto Poncelas describe ADAPT-DCU neural translation systems for the WMT Similar Language Translation task. The systems cover Croatian–Slovenian and Serbian–Slovenian in both directions.

  6. 2021

    Gender-balanced training data

    In 2021, she coauthors an automatic rewriting method for short conversational sentences. It uses neural machine translation to create gender-balanced outputs and training data.

  7. 2025

    First-person gender bias

    In 2025, she and Ekaterina Lapshinova-Koltunski present GENDER1PERSON, a test suite for bias in first-person singular translation from English into Russian and Serbian. Its dataset contains 1,000 Amazon product reviews. In 2025, she serves as an EMNLP workshop chair, with affiliations at Dublin City University and IU University in Berlin.

  8. 2026

    Lecturer and GITT committee member

    She is a lecturer at IU International University of Applied Sciences, a post she has held since 2023. In 2026, she serves on the program committee for the Gender-Inclusive Translation Technologies workshop, with an ADAPT Centre affiliation.

Career

  1. 2024–2025IU International University of Applied Sciences
  2. 2019–2023ADAPT - Centre for Digital Content Technology
  3. 2018–2025Dublin City University
  4. 2015–2018Humboldt-Universität zu Berlin
  5. 2011–2016DFKI

Education

  1. University of Belgrade
  2. RWTH Aachen UniversityPhD

From the biography and OpenAlex affiliations (the years of papers that list the organization), so some years are approximate.

Most-cited works

  1. chrF: character n-gram F-score for automatic MT evaluation2015 · Workshop on Statistical Machine Translation · 1,018 citations
  2. chrF++: words helping character n-grams2017 · Conference on Machine Translation (WMT) · 283 citations
  3. Towards Automatic Error Analysis of Machine Translation Output2011 · Computational Linguistics · 101 citationsWith Hermann Ney
  4. Word error rates2007 · Workshop on Statistical Machine Translation · 89 citationsWith Hermann Ney
  5. POS-based Word Reorderings for Statistical Machine Translation2006 · Language Resources and Evaluation Conference (LREC) · 88 citationsWith Hermann Ney

From OpenAlex, as of 21 October 2026.

  1. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Dublin City University. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  2. Nonprofit roles are 2.3 times as common in AI safetyIncludes DFKI. We estimate where people focused on safety, alignment and evaluation work. Nonprofit roles appear for 25.1% of safety people with known affiliations, compared with 11.0% in other areas.
  3. The top 10 organizations employed 22% of people shaping AI in 2025: 51% of company researchers and 21% of university researchersIncludes DFKI. We estimate the share of people shaping AI who worked at the top 10 organizations each year from 1955 to 2025, comparing concentration across companies, universities, and institutes and labs.
  4. More people shaping AI studied at Massachusetts Institute of Technology than at any other institutionIncludes RWTH Aachen University. We estimate where the people shaping AI studied and where their careers led. Massachusetts Institute of Technology leads with 467 former students, including 291 recorded doctorates.

Colleagues

People named in this biography, and people with dated positions at the same organization in the same years.

Sources and credits

Photo: We have no published photo of Maja Popović yet. Send us a free-licensed photo.

  1. Maja Popović · ADAPT, the Research Ireland Research Centre for AI-Driven Digital Content Technology adaptcentre.ie. Checked 21 October 2026.
  2. The RWTH Machine Translation System for WMT 2009 - ACL Anthology aclanthology.org. Checked 21 October 2026.
  3. Towards Automatic Error Analysis of Machine Translation Output - ACL Anthology aclanthology.org. Checked 21 October 2026.
  4. chrF: character n-gram F-score for automatic MT evaluation - ACL Anthology aclanthology.org. Checked 21 October 2026.
  5. Tools and Guidelines for Principled Machine Translation Development - ACL Anthology aclanthology.org. Checked 21 October 2026.
  6. Neural Machine Translation between similar South-Slavic languages - ACL Anthology aclanthology.org. Checked 21 October 2026.
  7. Generating Gender Augmented Data for NLP - ACL Anthology aclanthology.org. Checked 21 October 2026.
  8. GENDER1PERSON: Test Suite for Estimating Gender Bias of First-person Singular Forms - ACL Anthology aclanthology.org. Checked 21 October 2026.
  9. Organizing Committee - EMNLP 2025 2025.emnlp.org. Checked 21 October 2026.
  10. Maja Popovic (ADAPT) — Lacuna lacuna.tiptreesystems.com. Checked 21 October 2026.
  11. GITT 2026 sites.google.com. Checked 21 October 2026.