Shuly Wintner
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Research Interests
Shuly Wintner has a total of 80 co-authors. Here are the top co-authors by number of shared publications:
- Livnat Herzig Sheinfux
University of Haifa
4 shared publications - Noam Ordan
IAHLT, Jerusalem, Israel
4 shared publications - Nurit Melnik
The Open University of Israel
4 shared publications - Yulia Tsvetkov
University of Washington
4 shared publications - Doreen Osmelak
Department of Language Science and Technology, Universität des Saarlandes, Germany
3 shared publications - Ella Rabinovich
Department of Computer Science University of Haifa IBM Research † Language Technologies Institute Carnegie Mellon University
3 shared publications
Latest Publications
- Open Access
- 19 September 2025
- Published by Cambridge University Press (CUP) in Bilingualism: Language and Cognition
- Vol. 29 (4) , 1-15
Strategies of code-switching in human–machine dialogs
Most people are multilingual, and most multilinguals code-switch, yet the characteristics of code-switched language are not fully understood. We developed a chatbot capable of completing a Map Task with human participants using code-switched Spanish and English. In two experiments, we prompted the bot to code-switch according to different strategies, examining (1) the feasibility of such experiments for investigating bilingual language use and (2) whether participants would be sensitive to variations in discourse and grammatical patterns. Participants generally enjoyed code-switching with our bot as long as it produced predictable code-switching behavior; when code-switching was random or ungrammatical (as when producing unattested incongruent mixed-language noun phrases, such as ‘la fork’), participants enjoyed the task less and were less successful at completing it. These results underscore the potential downsides of deploying insufficiently developed multilingual language technology, while also illustrating the promise of such technology for conducting research on bilingual language use.
- 19 September 2025
- Published by Cambridge University Press (CUP) in Bilingualism: Language and Cognition
- Vol. 29 (4) , 1-15
Open AccessStrategies of code-switching in human–machine dialogs
Most people are multilingual, and most multilinguals code-switch, yet the characteristics of code-switched language are not fully understood. We developed a chatbot capable of completing a Map Task with human participants using code-switched Spanish and English. In two experiments, we prompted the bot to code-switch according to different strategies, examining (1) the feasibility of such experiments for investigating bilingual language use and (2) whether participants would be sensitive to variations in discourse and grammatical patterns. Participants generally enjoyed code-switching with our bot as long as it produced predictable code-switching behavior; when code-switching was random or ungrammatical (as when producing unattested incongruent mixed-language noun phrases, such as ‘la fork’), participants enjoyed the task less and were less successful at completing it. These results underscore the potential downsides of deploying insufficiently developed multilingual language technology, while also illustrating the promise of such technology for conducting research on bilingual language use.
- Open Access
- 28 May 2025
- Published by Springer Nature in Language Resources and Evaluation
- p. 1-32
The Knesset corpus: an annotated corpus of Hebrew parliamentary proceedings
We present the Knesset Corpus, a corpus of Hebrew parliamentary proceedings containing over 30 million sentences (over 384 million tokens) from all the (plenary and committee) protocols held in the Israeli parliament in the last three decades. Sentences are annotated with morpho-syntactic information and named entities, and are associated with detailed meta-information reflecting demographic and political properties of the speakers, based on a large database of parliament members and factions that we compiled. We discuss the structure and composition of the corpus and the various processing steps we applied to it. To demonstrate the utility of this novel dataset we present two use cases. We show that the corpus can be used to examine historical developments in the style of political discussions by showing a reduction in lexical richness in the proceedings over time. We also investigate some differences between the styles of male and female speakers. These use cases exemplify the potential of the corpus to shed light on important trends in the Israeli society, supporting research in linguistics, political science, communication, law, etc.
- 28 May 2025
- Published by Springer Nature in Language Resources and Evaluation
- p. 1-32
Open AccessThe Knesset corpus: an annotated corpus of Hebrew parliamentary proceedings
We present the Knesset Corpus, a corpus of Hebrew parliamentary proceedings containing over 30 million sentences (over 384 million tokens) from all the (plenary and committee) protocols held in the Israeli parliament in the last three decades. Sentences are annotated with morpho-syntactic information and named entities, and are associated with detailed meta-information reflecting demographic and political properties of the speakers, based on a large database of parliament members and factions that we compiled. We discuss the structure and composition of the corpus and the various processing steps we applied to it. To demonstrate the utility of this novel dataset we present two use cases. We show that the corpus can be used to examine historical developments in the style of political discussions by showing a reduction in lexical richness in the proceedings over time. We also investigate some differences between the styles of male and female speakers. These use cases exemplify the potential of the corpus to shed light on important trends in the Israeli society, supporting research in linguistics, political science, communication, law, etc.
- Open Access
Knesset-DictaBERT: A Hebrew Language Model for Parliamentary Proceedings
We present Knesset-DictaBERT, a large Hebrew language model fine-tuned on the Knesset Corpus, which comprises Israeli parliamentary proceedings. The model is based on the DictaBERT architecture and demonstrates significant improvements in understanding parliamentary language according to the MLM task. We provide a detailed evaluation of the model's performance, showing improvements in perplexity and accuracy over the baseline DictaBERT model.
Open AccessKnesset-DictaBERT: A Hebrew Language Model for Parliamentary Proceedings
We present Knesset-DictaBERT, a large Hebrew language model fine-tuned on the Knesset Corpus, which comprises Israeli parliamentary proceedings. The model is based on the DictaBERT architecture and demonstrates significant improvements in understanding parliamentary language according to the MLM task. We provide a detailed evaluation of the model's performance, showing improvements in perplexity and accuracy over the baseline DictaBERT model.
- Open Access
- 15 May 2024
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 59 (2) , 685-726
Analyzing learner language: the case of the Hebrew Learner Essay Corpus
We present the Hebrew Learner Essay Corpus (HELEECS): an annotated corpus of Hebrew language argumentative essays authored by prospective higher-education students. The corpus includes essays by two main populations: (1) essays by native speakers of Hebrew, written as part of the psychometric exam that is used to assess their future success in academic studies; (2) essays by non-native speakers of Hebrew, with three different native languages (Arabic, French, and Russian), that were written as part of a language aptitude test. The corpus is uniformly encoded and stored. The non-native essays were annotated with target hypotheses (i.e., hypothesized intended formulations in standard written Hebrew). The corpus is available for research purposes upon request. We describe the corpus and the error correction and annotation schemes used in its analysis. In addition to introducing this new resource, we discuss the challenges of identifying and analyzing non-native language use. Among these challenges are determining whether the language used in a particular utterance is native-like, and determining the target hypothesis when language use is non-native-like. We propose various ways for dealing with these challenges.
- 15 May 2024
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 59 (2) , 685-726
Open AccessAnalyzing learner language: the case of the Hebrew Learner Essay Corpus
We present the Hebrew Learner Essay Corpus (HELEECS): an annotated corpus of Hebrew language argumentative essays authored by prospective higher-education students. The corpus includes essays by two main populations: (1) essays by native speakers of Hebrew, written as part of the psychometric exam that is used to assess their future success in academic studies; (2) essays by non-native speakers of Hebrew, with three different native languages (Arabic, French, and Russian), that were written as part of a language aptitude test. The corpus is uniformly encoded and stored. The non-native essays were annotated with target hypotheses (i.e., hypothesized intended formulations in standard written Hebrew). The corpus is available for research purposes upon request. We describe the corpus and the error correction and annotation schemes used in its analysis. In addition to introducing this new resource, we discuss the challenges of identifying and analyzing non-native language use. Among these challenges are determining whether the language used in a particular utterance is native-like, and determining the target hypothesis when language use is non-native-like. We propose various ways for dealing with these challenges.
- Open Access
- 14 December 2023
- Published by MIT Press in Transactions of the Association for Computational Linguistics
- Vol. 11, 1471-1484
Shared Lexical Items as Triggers of Code Switching
Why do bilingual speakers code-switch (mix their two languages)? Among the several theories that attempt to explain this natural and ubiquitous phenomenon, the triggering hypothesis relates code-switching to the presence of lexical triggers, specifically cognates and proper names, adjacent to the switch point. We provide a fuller, more nuanced and refined exploration of the triggering hypothesis, based on five large datasets in three language pairs, reflecting both spoken and written bilingual interactions. Our results show that words that are assumed to reside in a mental lexicon shared by both languages indeed trigger code-switching, that the tendency to switch depends on the distance of the trigger from the switch point and on whether the trigger precedes or succeeds the switch, but not on the etymology of the trigger words. We thus provide strong, robust, evidence-based confirmation to several hypotheses on the relationships between lexical triggers and code-switching.
- 14 December 2023
- Published by MIT Press in Transactions of the Association for Computational Linguistics
- Vol. 11, 1471-1484
Open AccessShared Lexical Items as Triggers of Code Switching
Why do bilingual speakers code-switch (mix their two languages)? Among the several theories that attempt to explain this natural and ubiquitous phenomenon, the triggering hypothesis relates code-switching to the presence of lexical triggers, specifically cognates and proper names, adjacent to the switch point. We provide a fuller, more nuanced and refined exploration of the triggering hypothesis, based on five large datasets in three language pairs, reflecting both spoken and written bilingual interactions. Our results show that words that are assumed to reside in a mental lexicon shared by both languages indeed trigger code-switching, that the tendency to switch depends on the distance of the trigger from the switch point and on whether the trigger precedes or succeeds the switch, but not on the etymology of the trigger words. We thus provide strong, robust, evidence-based confirmation to several hypotheses on the relationships between lexical triggers and code-switching.
- Open Access
- 20 November 2023
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 59 (1) , 591-599
A corpus of English learners with Arabic and Hebrew backgrounds
- 20 November 2023
- Published by Springer Nature in Language Resources and Evaluation
- Vol. 59 (1) , 591-599
Open AccessA corpus of English learners with Arabic and Hebrew backgrounds
- Open Access
Shared Lexical Items as Triggers of Code Switching
Why do bilingual speakers code-switch (mix their two languages)? Among the several theories that attempt to explain this natural and ubiquitous phenomenon, the Triggering Hypothesis relates code-switching to the presence of lexical triggers, specifically cognates and proper names, adjacent to the switch point. We provide a fuller, more nuanced and refined exploration of the triggering hypothesis, based on five large datasets in three language pairs, reflecting both spoken and written bilingual interactions. Our results show that words that are assumed to reside in a mental lexicon shared by both languages indeed trigger code-switching; that the tendency to switch depends on the distance of the trigger from the switch point; and on whether the trigger precedes or succeeds the switch; but not on the etymology of the trigger words. We thus provide strong, robust, evidence-based confirmation to several hypotheses on the relationships between lexical triggers and code-switching.
Open AccessShared Lexical Items as Triggers of Code Switching
Why do bilingual speakers code-switch (mix their two languages)? Among the several theories that attempt to explain this natural and ubiquitous phenomenon, the Triggering Hypothesis relates code-switching to the presence of lexical triggers, specifically cognates and proper names, adjacent to the switch point. We provide a fuller, more nuanced and refined exploration of the triggering hypothesis, based on five large datasets in three language pairs, reflecting both spoken and written bilingual interactions. Our results show that words that are assumed to reside in a mental lexicon shared by both languages indeed trigger code-switching; that the tendency to switch depends on the distance of the trigger from the switch point; and on whether the trigger precedes or succeeds the switch; but not on the etymology of the trigger words. We thus provide strong, robust, evidence-based confirmation to several hypotheses on the relationships between lexical triggers and code-switching.
- Open Access
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 42-51
The Denglisch Corpus of German-English Code-Switching
Doreen Osmelak, Shuly Wintner. Proceedings of the 5th Workshop on Research in Computational Linguistic Typology and Multilingual NLP. 2023.
- 1 January 2023
- Published by Association for Computational Linguistics (ACL)
- p. 42-51
Open AccessThe Denglisch Corpus of German-English Code-Switching
Doreen Osmelak, Shuly Wintner. Proceedings of the 5th Workshop on Research in Computational Linguistic Typology and Multilingual NLP. 2023.
- Open Access
Machine Translation into Low-resource Language Varieties
State-of-the-art machine translation (MT) systems are typically trained to generate the "standard" target language; however, many languages have multiple varieties (regional varieties, dialects, sociolects, non-native varieties) that are different from the standard language. Such varieties are often low-resource, and hence do not benefit from contemporary NLP solutions, MT included. We propose a general framework to rapidly adapt MT systems to generate language varieties that are close to, but different from, the standard target language, using no parallel (source--variety) data. This also includes adaptation of MT systems to low-resource typologically-related target languages. We experiment with adapting an English--Russian MT system to generate Ukrainian and Belarusian, an English--Norwegian Bokm{\aa}l system to generate Nynorsk, and an English--Arabic system to generate four Arabic dialects, obtaining significant improvements over competitive baselines.
Open AccessMachine Translation into Low-resource Language Varieties
State-of-the-art machine translation (MT) systems are typically trained to generate the "standard" target language; however, many languages have multiple varieties (regional varieties, dialects, sociolects, non-native varieties) that are different from the standard language. Such varieties are often low-resource, and hence do not benefit from contemporary NLP solutions, MT included. We propose a general framework to rapidly adapt MT systems to generate language varieties that are close to, but different from, the standard target language, using no parallel (source--variety) data. This also includes adaptation of MT systems to low-resource typologically-related target languages. We experiment with adapting an English--Russian MT system to generate Ukrainian and Belarusian, an English--Norwegian Bokm{\aa}l system to generate Nynorsk, and an English--Arabic system to generate four Arabic dialects, obtaining significant improvements over competitive baselines.
- Open Access
Topics to Avoid: Demoting Latent Confounds in Text Classification
Despite impressive performance on many text classification tasks, deep neural networks tend to learn frequent superficial patterns that are specific to the training data and do not always generalize well. In this work, we observe this limitation with respect to the task of native language identification. We find that standard text classifiers which perform well on the test set end up learning topical features which are confounds of the prediction task (e.g., if the input text mentions Sweden, the classifier predicts that the author's native language is Swedish). We propose a method that represents the latent topical confounds and a model which "unlearns" confounding features by predicting both the label of the input text and the confound; but we train the two predictors adversarially in an alternating fashion to learn a text representation that predicts the correct label but is less prone to using information about the confound. We show that this model generalizes better and learns features that are indicative of the writing style rather than the content.
Open AccessTopics to Avoid: Demoting Latent Confounds in Text Classification
Despite impressive performance on many text classification tasks, deep neural networks tend to learn frequent superficial patterns that are specific to the training data and do not always generalize well. In this work, we observe this limitation with respect to the task of native language identification. We find that standard text classifiers which perform well on the test set end up learning topical features which are confounds of the prediction task (e.g., if the input text mentions Sweden, the classifier predicts that the author's native language is Swedish). We propose a method that represents the latent topical confounds and a model which "unlearns" confounding features by predicting both the label of the input text and the confound; but we train the two predictors adversarially in an alternating fashion to learn a text representation that predicts the correct label but is less prone to using information about the confound. We show that this model generalizes better and learns features that are indicative of the writing style rather than the content.