Anna Marklová

• ResearchGate
• Artificial Intelligence Language Lab
• Google Scholar
• ORCID

I am a corpus linguist and a psycholinguist, my research focuses mainly on the languages of artificial intelligence and AI slop. I completed my PhD at TU Dortmund and am currently based at the Faculty of Arts, Charles University, where I lead a project exploring linguistic variability in both human and AI language. Alongside the work on my own project, I also offer my skills as a painter and illustrator, creating visual stimuli for linguistic experiments.

What do people intuitively know about registers (the way language shifts between a pub conversation and a scientific paper) and does that intuition match what corpora tell us? And can large language models do the same? This project combines corpus and psycholinguistic evidence on register variation in various languages, with a growing focus on machine-generated language(s). The first corpus of Czech LLM-generated texts is being built as part of the project, tracking how models (and the language they produce) change over time, using stylometry, register analysis, and other methods.

The project is funded from 2025 to 2028.

Me and Jiří Milička founded the Artificial Intelligence Language Lab, which brings together researchers at the Faculty of Arts and beyond who study the languages of AI.

• PRIMUS: Sensitivity towards register-specific features: Combination of corpus and experimental evidence
• GAČR: Large language models through the prism of corpus linguistics

Department of Linguistics, Faculty of Arts, Charles University

Principal investigator of the PRIMUS project Sensitivity towards register-specific features: Combination of corpus and experimental evidence

Institute of the Czech National Corpus, Faculty of Arts, Charles University

Member of the GAČR project Large language models through the prism of corpus linguistics

Humboldt-Universität zu Berlin, Humboldt-Innovation

Conception and creation of online materials, testing and linguistic evaluation — responsible for the linguistic evaluation and analysis.

Seminar für Slavistik / Lotman-Institut, Ruhr-Universität Bochum

Project Animacy marking in Upper Sorbian and Polish in Germany - designing and creating the picture stimuli for the main experiments.

Sprach- und literaturwissenschaftliche Fakultät, Humboldt-Universität zu Berlin

Researcher in the CRC 1412 Register project Expressive dislocation and register in Czech vs. Russian, applying psycholinguistic methods to the results of a multidimensional analysis of Czech corpora.

Fakultät Kulturwissenschaften, Technische Universität Dortmund

Work in the psycholinguistics laboratories, including eye-tracking and speech elicitation experiments for the dissertation project The Expression of Endpoints in Early Child's Visual Perception and Speech; teaching duties, leading university courses and supervising students' theses.

Institute of the Czech National Corpus, Prague

Undergraduate assistant in corpus linguistic research.

• Marklová, A.; Vinš, O.; Vokáčová, M.; Milička, J. (2026). The author is dead, but what if they never lived? A reception experiment on Czech AI- and human-authored poetry. Digital Scholarship in the Humanities, fqag067. DOI
• Delucchi Danhier, R.; Marklová, A.; Meese, R.; Griebenow, L. (2026). From symmetry to asymmetry in sentence representation: the role of literacy and syntax in spatial bias development. Reading and Writing. DOI
• Marklová, A.; Meyer, R.; Szucsich, L.; Buchmüller, O.; Demian, C. (accepted for publication). Register-based preferences of long topicalization constructions in Czech: A forced-choice study. Advances in Formal Slavic Linguistics 2023.
• Marklová, A.; Milička, J.; Vokáčová, M.; Rosa, R. (2026). Production and perception in LLMs: A token probability approach. arXiv. preprint
• Marklová, A.; Milička, J.; Vokáčová, M. (accepted for publication). The problem of including experts as participants in empirical research: Evidence from linguistics. International Journal of Social Research Methodology.
• Milička, J.; Marklová, A.; Cvrček, V. (accepted for publication). AI Brown and AI Koditex: LLM-generated corpora comparable to human-written corpora of English and Czech texts. Language Resources and Evaluation. preprint
• Milička, J.; Marklová, A.; Cvrček, V. (accepted for publication). Benchmark of stylistic variation in LLM-generated texts. International Journal of Corpus Linguistics. preprint

• Marklová, A.; Demian, C.; Meyer, R.; Szucsich, L.; Buchmüller, O. (2025). Sensitivity towards registers in Czech. Register Studies. DOI
• Marklová, A.; Milička, J.; Ryvkin, L.; Lacková Bennet, L.; Kormaníková, L. (2025). Iconicity in large language models. Digital Scholarship in the Humanities, fqaf095. DOI
• Marklová, A.; Delucchi Danhier, R. (2025). Human asymmetries in AI art: Syntax and writing direction effects on agent position in AI-generated images. PLOS ONE, 20(11), 1–19. DOI
• Milička, J.; Marklová, A.; Drobil, O.; Pospíšilová, E. (2025). Learning to detect AI texts and learning the limits. PLOS ONE, 20(10), e0333007. DOI
• Milička, J.; Marklová, A.; Láznička, M.; et al. (2025). Sources of intelligibility of distant languages: An empirical study. Language and Speech. DOI
• Křen, M.; Cvrček, V.; Čapka, T.; Hnátková, M.; Jelínek, T.; Kocek, J.; Kováříková, D.; Křivan, J.; Marklová, A.; Petkevič, V.; Skoumalová, H.; Škrabal, M. (2025). SYN2025: reprezentativní korpus psané češtiny. Ústav Českého národního korpusu FF UK, Praha. korpus.cz

• Marklová, A.; Houzar, A.; Meyer, R.; Buchmüller, O.; Demian, C.; Szucsich, L. (2024). Left dislocations and long topicalizations in Czech: An acceptability judgements study. Zeitschrift für Slawistik, 69(3), 553–574. DOI
• Milička, J.; Marklová, A.; VanSlambrouck, K.; et al. (2024). Large language models are able to downplay their cognitive abilities to fit the persona they simulate. PLOS ONE, 19(3), e0298522. DOI
• Marklová, A.; Panfilova, E.; Mertins, B. (2023). Description of locomotion events: Language-specific patterns in child-directed speech. WORD, 69(4), 339–361. DOI
• Marklová, A.; Mertins, B. (2023). Perception of goal-oriented locomotion events in monolingual and bilingual adults. In B. Mertins & R. Delucchi Danhier (Eds.), Diversity in Cognition (Vol. 20, pp. 91–118). Peter Lang. DOI
• Marklová, A.; Buchmüller, O.; Demian, C.; Meyer, R.; Szucsich, L. (2023). Register in Czech: Designing an MDA-based experimental study. Linguistics Beyond and Within, 9, 60–75. DOI
• Marklová, A.; Delucchi Danhier, R. (2023). When spatial agency bias and the advantage of the first mention are in contradiction: Evidence from Czech, German, and Spanish. Linguistics Beyond and Within, 9, 76–94. DOI
• Marklová, A.; Mertins, B. (2023). Expression and perception of endpoints during language acquisition: Three studies on Czech. Studies in Applied Linguistics, 14(2), 7–29.
• Marklová, A. (2023). Investigating motion events: Endpoints in speech and perception of children and adults [Dissertation]. TU Dortmund. DOI

• Marklová, A.; Panfilova, E.; Mertins, B. (2023). Parent-child conversations about motion events (Russian, Russian-German, Czech). DataverseNO, V1. DOI

• Marklová, A.; Vinš, O.; Vokáčová, M.; Milička, J. (2026). The author is dead, but what if they never lived? Reception of AI-generated poetry in Czech. Psychoslav 2026, Zadar, Croatia.
• Marklová, A.; Delucchi Danhier, R. (2026). Multimodal corpus of AI-generated images: Investigating linguistic bias in English-based visual models. ICAME 47, Koblenz, Germany.
• Tichý, O.; Bulantová, B.; Titlbachová, M.; Marklová, A.; Milička, J. (2026). Benchmarking large language models for linguistic research. ICAME 47, Koblenz, Germany.
• Marklová, A.; Delucchi Danhier, R. (2025). Reading-writing direction effects in human and AI visual conceptualization. CALC-3, Brno, Czech Republic.
• Pospíšilová, E.; Drobil, O.; Marklová, A.; Milička, J. (2025). Humans are bad at recognizing AI — but they can learn it from feedback. AMLaP, Prague, Czech Republic.
• Pospíšilová, E.; Drobil, O.; Marklová, A.; Milička, J. (2025). Humans are bad at recognizing AI — but they can learn it from feedback. Summer School of Linguistics, České Budějovice, Czech Republic.
• Milička, J.; Marklová, A.; Cvrček, V. (2025). AI-Brown Corpus. Corpus Linguistics 2025, Birmingham, United Kingdom.
• Marklová, A.; Milička, J.; Pospíšilová, E.; Drobil, O. (2025). The future of corpus linguistics in the world of large language models. ICAME, Vilnius, Lithuania.
• Pospíšilová, E.; Drobil, O.; Marklová, A.; Milička, J. (2025). Známe umělou inteligenci lépe, než ona zná nás? Vliv zpětné vazby na rozpoznávání textů generovaných AI. UIVID, Plzeň, Czech Republic.
• Drobil, O.; Pospíšilová, E.; Marklová, A.; Milička, J. (2025). Do we know AI better than it knows us? The effect of feedback on detecting AI-generated texts. YLMP, Poznań, Poland.
• Marklová, A.; Milička, J.; Ryvkin, L. (2025). Large language models' ability to generate iconic pseudowords. Workshop on Dimensions of Iconicity in the Visual Modality, University of Göttingen, Germany.
• Milička, J.; Marklová, A. (2025). Register variability in machine-generated texts. ISCA/ITG Workshop, Berlin, Germany.
• Marklová, A.; Milička, J.; Ryvkin, L.; Lacková, L.; Kormaníková, L. (2025). Large language models' ability to generate iconic pseudowords. Bienále české lingvistiky 2025, Prague, Czech Republic.
• Milička, J.; Marklová, A.; Cvrček, V. (2024). Exploring register variation in human and machine-generated texts: A comparative analysis. ICAME, Vigo, Spain.
• Marklová, A.; Delucchi Danhier, R.; Griebenow, L. (2023). Development of reading-writing direction effects on spatial object positioning: Insight into German pupils. 6th Białystok-Kyiv Conference on Theoretical and Applied Linguistics, Białystok, Poland.
• Marklová, A.; Meyer, R.; Buchmüller, O.; Demian, C.; Szucsich, L. (2023). Register-based acceptability of left dislocation and long topicalization in Czech. Formal Description of Slavic Languages 16, Graz, Austria.
• Delucchi Danhier, R.; Vidua, B.; Marklová, A. (2023). Differences in prototypes can be attributed to culture: Examples from German, Spanish and Czech. Linguistics Beyond and Within 2023, Lublin, Poland.
• Marklová, A.; Buchmüller, O.; Demian, C.; Meyer, R.; Szucsich, L. (2023). Left dislocation and long topicalization in Czech: An acceptability study. Summer School of Linguistics 2023, České Budějovice, Czech Republic.
• Marklová, A. (2023). Expression and perception of endpoints during language acquisition: Three studies on Czech. Psycholinguistics in Flanders 2023, Ghent, Belgium.
• Marklová, A.; Buchmüller, O. (2022). Register investigation on Czech: Designing an MDA-based experimental study. Linguistic SFB Meetup 2022, Berlin, Germany.
• Marklová, A.; Schnelle, G.; Lehman, N. (2022). Reading circle terminology. CRC 1412 Spring Retreat 2022, Berlin, Germany.
• Marklová, A.; Buchmüller, O.; Demian, C.; Meyer, R.; Szucsich, L. (2022). Register investigation on Czech. Summer School of Linguistics 2022, České Budějovice, Czech Republic.
• Marklová, A.; Delucchi Danhier, R. (2022). Spatial asymmetry of mental representation in Czech, German and Spanish. Deutscher Slavistiktag 2022, Bochum, Germany.
• Marklová, A.; Buchmüller, O.; Demian, C.; Meyer, R.; Szucsich, L. (2022). Speakers' assessment of situational contexts of language use: Dimensions of variation in Czech. Linguistics Beyond and Within 2022, Lublin, Poland.
• Marklová, A.; Delucchi Danhier, R.; Meese, R. (2022). Spatial agency bias and the advantage of first mentioned are learned: (Negative) evidence from German pre-schoolers. Linguistics Beyond and Within 2022, Lublin, Poland.
• Szucsich, L.; Meyer, R.; Buchmüller, O.; Marklová, A.; Demian, C. (2021). Expressive dislocation and register in Czech vs. Russian. CRC 1412 Autumn Retreat 2021, Berlin, Germany.
• Marklová, A. (2021). Motion events in the light of linguistic relativity. Kolloquium zur Mehrsprachigkeitsforschung, Bochum, Germany.
• Marklová, A.; Panfilova, E.; Mertins, B. (2020). The expression of endpoints in child-directed speech of Czech, Russian and Russian-German parents: A cross-linguistic study. Psycholinguistics in Flanders 2020, Kaiserslautern, Germany.
• Marklová, A.; Mertins, B. (2019). The effect of grammar on perception and verbalization of locomotion events in monolingual and bilingual children and adults. International Symposium of Bilingualism, Edmonton, Canada.
• Marklová, A. (2019). Seeing for speaking: Eye-tracking in linguistic research. Interdisciplinary PhD Colloquium, Dortmund, Germany.
• Marklová, A.; Mertins, B. (2018). Endpoint preferences in bilingual infants. International Symposium on Bilingual Processing in Adults and Children, Braunschweig, Germany.

• Marklová, A. (2026). AI slop through a psycholinguistic lens. Seminar of Formal Linguistics, Institute of Formal and Applied Linguistics, Faculty of Mathematics and Physics, Charles University, Prague, Czech Republic.
• Marklová, A. (2021). Vidět jazykem: Eye-tracking v lingvistickém výzkumu. Evropský den jazyků 2021, Prague, Czech Republic.
• Panelist at the colloquium Towards AI-Aided Human-Supervised Linguistics (2023), Prague, Czech Republic.

Consider the environment before printing this web page.