Sameer Singh

Career

Sameer Singh received a PhD from the University of Massachusetts Amherst, with internships at Microsoft Research, Google Research and Yahoo! Labs, and was a postdoctoral researcher at the University of Washington from 2013 to 2016. Singh joined UC Irvine as an assistant professor of computer science in 2016, became an associate professor in 2021 and a full professor in 2024, with affiliations in Linguistics and EECS.¹

From 2021 to 2023 Singh was an Allen Fellow at the Allen Institute for AI, and Singh's site lists a current role as co-founder and CTO of Spiffy AI.¹

Research

Singh's research concerns the robustness and interpretability of machine learning and models that reason over text and structure. LIME, introduced in 2016 with Marco Tulio Ribeiro and Carlos Guestrin, explains an individual prediction of any classifier by fitting an interpretable model locally around it. A related 2016 paper on model-agnostic interpretability received the workshop best paper award at the ICML Workshop on Human Interpretability in Machine Learning.¹²⁴

CheckList (2020), again with Ribeiro and others, proposed behavioral testing of NLP models with a matrix of linguistic capabilities and test types, and received the best paper award at ACL 2020. Later award-winning work from Singh's group includes FRUIT, a NAACL 2022 task on updating text to reflect new information.³⁴

Work and contributions

• LIME (Why Should I Trust You? Explaining the Predictions of Any Classifier) Published research

  Co-author of the 2016 paper proposing LIME, which explains an individual prediction by learning an interpretable model locally around it.²

• Beyond Accuracy: Behavioral Testing of NLP models with CheckList Published research

  Co-author with Marco Tulio Ribeiro, Tongshuang Wu, and Carlos Guestrin of the 2020 CheckList paper on behavioral testing of NLP models.³

Timeline

1. 2013Began a postdoctoral position at the University of Washington.¹
2. 2016Joined UC Irvine as an assistant professor of computer science.¹
3. Feb 2016Co-authored the LIME paper on explaining the predictions of any classifier.²
4. May 2020Co-authored CheckList, a behavioral testing method for NLP models.³
5. 2021Became an associate professor at UC Irvine and an Allen Fellow at the Allen Institute for AI.¹
6. 2024Became a professor of computer science at UC Irvine.¹

Other documented roles

Roles the sources document that are not confirmed as current.

• Allen Fellow (2021–2023), Allen Institute for AI¹
• Postdoctoral Researcher (2013–2016), University of Washington¹

Honors and recognition

• DARPA Riser · DARPA, 2015⁴
• Best Paper Award, ACL 2020, for CheckList · Association for Computational Linguistics, 2020⁴
• Hellman Fellowship · Hellman Family Foundation, 2020⁴
• NSF CAREER Award · U.S. National Science Foundation, 2021⁴
• Best Task Paper Award, NAACL 2022, for FRUIT · North American Chapter of the Association for Computational Linguistics, 2022⁴
• Kavli Fellow · National Academy of Sciences (Kavli Frontiers of Science), 2023⁴

Where they work

• UC Irvine 1 person in the wiki
• Spiffy AI 1 person in the wiki

Related people

• Thomas ScialomLanguage models · Evaluation and benchmarks
• Ian R. McKenzieEvaluation and benchmarks · Language models
• Emmanuel DupouxLanguage models · Evaluation and benchmarks
• Yulia TsvetkovLanguage models · Evaluation and benchmarks
• Margaret MitchellEvaluation and benchmarks · Language models
• Sarthak JainLanguage models · Interpretability
• Samuel R. BowmanEvaluation and benchmarks · Language models
• Kate SouleLanguage models · Evaluation and benchmarks
• Su Lin BlodgettEvaluation and benchmarks · Language models
• Ethan PerezEvaluation and benchmarks · Language models