Image

About me

I'm Christin Beck (née Schätzle), a postdoc at the Department of Linguistics, University of Konstanz, where I am a member of the Computational Linguistics group. I am also part of the collaborative research center SFB-TRR 161 'Quantitative Methods for Visual Computing', where I develop novel and innovative visualization methods for the analysis of multidimensional language data in collaboration with experts from the field of Visual Analytics. Broadly, my research focuses on the development of computational and corpus analytic methods for the investigation of historical language change. I am further interested in computational lexical semantics, the interaction between lexical semantic and syntactic change and the syntactic encoding of information structure. More recently, I began investigating how contextualized language models can be leveraged for historical linguistic research. With respect to this, I have explored how function words are captured by contextualized language models such as BERT.

News and upcoming events

Fellowship 'Words, Bones, Genes, Tools' March-June 2023
From March-June 2023, I will join the DFG Center for Advanced Studies 'Words, Bones, Genes, Tools' at the University of Tübingen for a four month fellowship. During the fellowship, I will collaborate with Marisa Köllner, working on developing a new methodology for investigating lexical semantic change across language stages which combines the usage of contextualized language models and phylogenetic methodologies for cognate identification. More on the project can be found here.
Paper at ACL-IJCNLP, August 2021
Together with Rita Sevastjanova, Aikaterini-Lida Kalouli, Hannah Hauptmann (née Schäfer) and Mennatallah El-Assady, I worked on a paper on 'Explaining Contextualization in Language Models using Visual Analytics', where we develop a visual analytics and scoring technique for examining contextualization in word embeddings across the different layers which are part of the language model (i.e., in our case BERT-base with 12 layers). Our paper was accepted for presentation at the joint conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021) and will be published as part of the proceedings of the conference in August 2021.
12th International Conference of Nordic and General Linguistics (NGL12), June 2021
Hannah Booth and I will give a talk on 'V1, V2 and information structure in the history of Icelandic' at the 12th International Conference of Nordic and General Linguistics (NGL12), taking place online (organized by the University of Oslo), on June 18th.
Diachronic Generative Syntax (DiGS) conference 22, May 2021
I have two presentations at DiGS22, which is taking place online on 20th-22nd May 2021. Together with Hannah Booth, I will give a talk about Information structure and word order change: verb-first and verb-second in Icelandic. With Ashwini Deo and Miriam Butt, I will present work on Tracking Case Innovation: A Perspective from Marathi.
Paper in Journal of Historical Syntax, May 2021
Hannah Booth and I just published a paper on 'Verb-second and verb-first in the history of Icelandic' in the Journal of Historical Syntax (JHS; open access).