Job Openings

Collaborative Research Center (CRC) 1102 / Sonderforschungsbereich (SFB) 1102

“Information Density and Linguistic Encoding”
Saarland University, Germany

3 Post-doctoral and 11 doctoral positions available


The DFG-funded CRC in Information Density and Linguistic Encoding (SFB 1102) is pleased to invite applications for a range of post-doctoral and doctoral positions. The overarching research question the CRC addresses is to what extent Information Theory can contribute to a unifying model of language use, variation and change. Against this background, language is viewed from the perspective of (bounded) rational communication, according to which interlocutors strive to modulate the encoding of their messages so as to (i) successfully convey their intended message, and (ii) optimize their cognitive effort.

The CRC includes 17 research projects drawing upon psycholinguistics/neurolinguistics, computational linguistics, diachronic sociolinguistics, phonetics, discourse and contrastive linguistics and translatology with their respective empirical methods, ranging from computational language modeling to experimental and corpus-based methods. We are seeking to recruit 3 post-doctoral and 11 doctoral students.


  • starting date: from September 1, 2022
  • application deadline: June 30, 2022
  • interviews: starting July 1

Reference number W2125, salary in accordance with the German TV-L salary scale, pay grade: E13, employment: 75% or 100 % of standard working time. The employment relationship varies for the duration of the relevant qualification, at the latest until June 30, 2026.


SFB 1102, Information Density and Linguistic Encoding

Job requirements and responsibilities:

  • Research as doctoral student or postdoc in the following fields: psycholinguistics/neurolinguistics, computational linguistics, diachronic sociolinguistics, phonetics, discourse and contrastive linguistics, language typology and translatology
  • For detailed project descriptions and requirements see:

Your academic qualifications:

  • Employment requirements are PhD for postdoctoral positions or MA/MSc for doctoral positions. Note that we are happy to receive applications by people who have not yet finished their MA/MSc or PhD by the time of application but will have submitted their thesis by the starting date or shortly thereafter.

We look forward to receiving your meaningful online application (in a PDF file) by 30.06.2022 to Please include „Job application CRC 1102 /  W2125″ in the subject line of the e-mail. Applicants are requested to submit their application, together with an academic CV, a list of academic publications, copies of academic degree certificates and two potential references.

Please indicate which project(s) you wish to apply to by stating the project number(s). We also reserve the right to forward applications to other projects if suitable. Applications received by June 30 will be given full consideration, while those applications received after June 30 will be considered until positions are filled.

If you have any questions, please contact us for assistance. Your contact:

SFB coordination team

For detailed project descriptions and requirements see:

Pay grade classification is based on the particular details of the position held and the extent to which the applicant meets the requirements of the pay grade within the TV-L salary scale. Part-time employment is generally possible.

If you have obtained a foreign university degree, a proof of the equivalence of this degree with a German degree by the Zentralstelle für ausländisches Bildungswesen (ZAB) is needed before hiring. If necessary, please apply for this in time. You can find more information at

Unfortunately, neither costs for attending an interview at Saarland University nor costs for any certificate evaluation by the ZAB can be reimbursed in principle.

We welcome applications regardless of gender, nationality, ethnic and social origin, religion/belief, disability, age, and sexual orientation and identity. In accordance with its policy of increasing the proportion of women, the University actively encourages applications from women. Applications from severely disabled persons will be given preferential consideration in the event of equal suitability.

When you submit a job application to Saarland University you will be transmitting personal data. Please refer to our privacy notice for information on how we collect and process personal data in accordance with Art. 13 of the Datenschutz-Grundverordnung. By submitting your application you confirm that you have taken note of the information in the Saarland University privacy notice.


Details on the vacant positions can be found below, sorted by project:


Project A5: The Role of Language Experience and Surprisal in Learning and Memory

PIs: Jutta Kray & Katja Häuser

The main objective of project A5 is to examine whether the processing of confirmed and disconfirmed predictions (surprisal) during language comprehension influence the learning of novel word meanings and the retrieving of information from long-term memory as a function of language experience. In order to investigate the influence of language experience, we will use a lifespan approach and mainly compare age groups differing in their lifetime exposure to language, specifically, children, younger and older adults. The impact of confirmed and disconfirmed predictions (surprisal) on language comprehension will be assessed by means of behavioral, eye-tracking and EEG methods.


Doctoral Researcher (TV-L 13, 75%) – The successful candidate should have a background in developmental psychology, experience with testing children with eye-tracking or EEG methods, advanced knowledge in data analysis with R, as well as some expertise in programming lab or online experiments.

Project A6: Expectancy-based mechanisms during language comprehension and their relation to memory formation and retrieval

PIs: Axel Mecklinger & Regine Bader

Project A6 uses behavioural and electrophysiological measures to investigate how expectancy-based mechanisms in natural language contexts modulate learning and memory processes. In detail, we explore how predictive processing shapes memory processes in different learning situations, i.e., artificial grammar learning, the acquisition of new semantic knowledge, multilingual vocabulary learning and event segmentation in narrative reading.


Doctoral Researcher (TV-L 13, 75%) – The successful candidate should have a very good Master degree in psychology, neuroscience or a related program. Applicants with a strong interest in neuroscientific research on learning, memory and language, ample experimental and statistical knowledge, and a very good command of English will be given preference. Major responsibilities include planning and running behavioral and EEG experiments, data analysis and contribution to publications and project reports. Familiarity with electrophysiological methods as well as programming and analysis skills (e.g., Matlab, PsychoPy, E-Prime, R) would be an advantage. The position holder is expected to prepare a PhD Dissertation in this domain.

Project A7: Controlling Information Density in Discourse Generation

PIs: Jörg Hoffmann & Alexander Koller

The  goal of this research project is to generate effective building  instructions in Minecraft. We already have a system which uses  techniques from natural language generation and AI planning to plan  sequences of instructions at an optimal level of abstraction (can we say  „build a railing“ or do we have to explain this block by block?). Now  we want to make the NLG system automatically adapt to the user by  identifying user groups in unannotated training data. At the same time,  we will extend our symbolic NLG system into a neurosymbolic one.


Postdoctoral  Researcher (TV-L 13, 100%) – The successful candidate should have a PhD in  Computational Linguistics, Computer Science, or a related discipline,  with a strong background in neural or neurosymbolic models of language  and symbolic parsing or generation algorithms. Strong programming skills  and a good command of English are essential.

Project A8: Adapting text generation to individual users

PIs: Vera Demberg & Alexander Koller

In this  research project, we will develop a system which automatically rewrites  text to be optimally readable for an individual user. This project  brings together psycholinguistic and NLP research: we will identify how a  text should be rewritten to be optimal for a user from their eye  movements while reading, and we will then apply neural text-to-text  generation methods to rewrite the text with respect to lexical and  syntactic complexity and the prior knowledge the text assumes.


Doctoral Researcher (TV-L 13, 75%) – We are looking for a PhD student to do research  on the NLP part of the project. The successful candidate should have an  MSc degree in Computational Linguistics, Computer Science, or a related  discipline, with a demonstrated interest in NLP (especially neural  models) and an interest in carrying out interdisciplinary research.  Strong programming skills and a good command of English are essential.


Project B1: Information Density in English Scientific Writing: A Diachronic Perspective

PIs: Elke Teich & Stefania Degaetano-Ortlieb

Project B1 is concerned with the hypothesis of communicative optimization in the development of scientific English. Based on a diachronic corpus (Royal Society Corpus), we have applied selected types of computational language models (e.g. topic models, n-gram models, word embeddings) and combined them with information-based measures (e.g. entropy, surprisal) to capture diachronic variation.

In the upcoming project phase we intend to address the following research questions: (1) Is there evidence of a more general diachronic mechanism (beyond scientific language)? (2) Within scientific language, what are additional, typical imprints of conventionalization (register variables)? (3) How can we assess the overall communicative efficiency of scientific language (effects on working memory)? Focusing on selected linguistic phenomena (multi-word expressions, discourse markers, nominal vs. verbal phrases), we complement a corpus-based, production-oriented approach with selected behavioural, comprehension-oriented studies.


Doctoral Researcher (TV-L 13, 75%) – The successful candidate should have a background in corpus and experimental linguistics and solid computational skills. A good command of English is mandatory. Working knowledge of German is desirable.

Project B3: Information Density and Ellipsis Redundancy

PIs: Ingo Reich, Heiner Drenhaus, Robin Lemke

Project  B3 investigates why and under which circumstances speakers use ellipses  in coordinations. Building on information theoretical concepts, the  project focuses on predictability effects driven by both linguistic and  extralinguistic context, interactions between speaker and hearer as well  as memory limitations, which might determine the choice between  elliptical and non-elliptical utterances. To investigate the impact of  these factors on ellipsis production and comprehension, B3 uses  self-paced reading, production and acceptability rating experiments  (both web- and lab-based) and eye-tracking.


Doctoral Researcher(TV-L 13,  75%) – The successful candidate should have a Master’s degree either in  psycholinguistics or in theoretical linguistics, and a good background  in the other area, respectively. The PhD student’s tasks include the  planning and conduction of experiments as well as the statistical  analysis and presentation of the experimental results. The student is  also expected to prepare a PhD thesis in this domain. Since the  experiments will be conducted in German, good command of German is a  prerequisite. Experience with statistical methods (in particular mixed  effects modeling) and/or programming skills are desirable.

Projekt B4: Modelling and Measuring Information Density

PI: Dietrich Klakow

This research project is concerned with the analysis and improvement of neural language models. It studies the adaptation of language models as well as how to use them for dynamically changing tasks and data distributions. We are looking for a PhD candidate that will work on developing language models with different memory capacities in order to explore “cognitively plausible” language models.


Doctoral Researcher (TV-L 13, 75%) – The successful candidate should have a background in computational linguistics, computer science or a related subject, ideally with experience in natural language processing. The candidate is expected to prepare a PhD dissertation in this domain.

Project B6: Unraveling Linguistic Knowledge via Multi-lingual Embedding Spaces and
Latent Information

PIs: Josef van Genabith, Cristina España i Bonet

The research project is concerned with multilinguality and translationese addressing both theoretical and practical questions. The project explores information spreading in multilingual embedding spaces, analysis and removal of translationese embedding subspaces in NLP and extracting latent background knowledge from bilingual data. We seek to apply answers to the foundational questions to improve NLP applications, including multilingual NLP for low-resource languages.


Postdoctoral Researcher (TV-L 13, 100%) – The successful candidate should have a Ph.D. in Computer Science, Computational Linguistics, or a related discipline, with a strong background in machine learning and natural language processing, preferably in machine translation, natural language understanding, multilingual technologies, and deep learning. A strong track record and programming skills are essential. We expect excellent problem solving skills, independent and creative thinking, excellent team working and communication skills, and a good command of written and oral English. The responsibilities include publication in top-tier conferences and journals, and contribution to teaching and supervision.


Doctoral Researcher (TV-L 13, 75%) – The successful candidate should have a master in Computer Science, Computational Linguistics, or a related discipline, with a strong background in natural language processing, preferably in machine translation or multilingual technologies. Strong programming skills are essential. A good command of English is mandatory. We expect excellent problem solving skills, independent and creative thinking, excellent team working and communication skills, and a good command of written and oral English. The candidate is expected to prepare a PhD dissertation in this domain and publish in top-tier conferences.

Project B7: Translation as Rational Communication

PIs: Elke Teich & Ekaterina Lapshinova.Koltunski 

Project B7 focuses on the specific linguistic properties of translation, i.e. non-randomly-occurring linguistic features that distinguish translations from original productions. Such properties are commonly referred to as “translationese” and emerge through the translation-inherent dilemma of ensuring source language fidelity while adhering to target language rules and norms. Our overarching research question is to what extent translationese effects can be described and explained in an information-theoretic framework of rational communication. Our approach is corpus-based using selected computational language models and information-theoretic measures including surprisal and entropy to assess translationese effects. Adopting an information-theoretic perspective on translationese is a novel idea and promises new insights into the general mechanisms underlying translation.

Postdoctoral Researcher (TV-L 13, 100%), computational linguist / computer scientist.

Task(s). Analysis of computational language models (comparable and parallel corpora, languages: English, German, Spanish) for the linguistic study of translationese


  • Expertise in computational language modeling, especially multilingual language modeling (e.g. multilingual word embeddings), surprisal and memory-surprisal
  • Experience in working with small data sets
  • Lexical aspects: working with bilingual lexicons, knowledge of multilingual lexical semantics, cross-lingual lexical databases (e.g. WordNets)
  • Syntactic aspects: experience with universal dependencies, using multilingual parsed data for language modeling
  • Discourse: experience with computational approaches to cross-lingual analysis of discourse connectives and coreference
  • Spoken data: knowledge in processing various data formats (audio, time-aligned data, text-aligned data).

Knowledge of German is desirable, knowledge of Spanish is an advantage.

Project C3: Rational Encoding and Decoding of Referring Expressions

PIs: Matthew Crocker , Heiner Drenhaus &  Noortje Venhuizen

Project C3 investigates the extent to which rational communicative strategies can explain speakers‘ production choices and the consequences for listeners, using controlled neuro-behavioral experiments and computational modelling. The project examines the relation between information theory and information structure in online comprehension, as well as the interaction of linguistic encoding and decoding with the visual situational context.

Doctoral Researcher (TV-L 13, 75%), Psycholinguist – The doctoral student should have a masters degree in psycholinguistics,or a related program, with knowledge of behavioral and/or ERP methods, and statistical analysis. The focus of this position is on ERP experiments related to information structure and information theory. A background in theoretical linguistics is an advantage. Good knowledge of German is strongly desired.


Doctoral Researcher (TV-L 13, 75%) – Cognitive modeller. The doctoral student should have a masters degree in cognitive science, or a related programme, with experience in connectionist modelling and/or computational psycholinguistics. The focus of this position will be on developing cognitive models for human language comprehension and production, complemented by psycholinguistic experimentation.

Project C7: Cross-Linguistic Information-Theoretic Modelling of Communicative Effciency

PI: Annemarie Verkerk

The C7 project is concerned with the impact of information structure on information-theoretic modelling of word order variability across a sample of 40+ languages. Information-theoretic modelling will be done in terms of dependency grammar, i.e. measuring dependency locality and calculating the memory-surprisal tradeoff. The data source is a parallel corpus, which has to be annotated for information status using a mixture of crowd-sourcing and computational methods. C7 then considers which word orders are variable across languages, and how word order variability interacts with information status and minimization of dependencies, both from a cross-linguistic as well as a language-internal perspective.


Doctoral Researcher (TV-L 13, 75%) – The successful candidate should have a strong background in computational linguistics. A good command of English is mandatory. Knowledge of a non-European language and/or corpus linguistics is desirable. The successful candidate should be prepared to prepare a PhD dissertation on the incorporation of information status in information-theoretic modelling using a combination of computational and experimental approaches.

Project T1: Information Density and Linguistic Encoding in „Leichte Sprache“ (IDeaLite)

PIs: Ingo Reich, Heike Zinsmeister, Elke Teich

In  enger Zusammenarbeit mit unserem Kooperationspartner AWO Saarland  untersucht Projekt T1 auf dem Hintergrund informationstheoretischer  Konzepte den Gebrauch und die Verarbeitung Leichter Sprache. Das Projekt  nutzt dabei sowohl korpuslinguistische wie auch experimentelle  Methoden. Die Anwendung informationstheoretischer Modelle auf Korpora  Leichter Sprache soll einerseits die Strukturen des tatsächlichen  Gebrauchs offen legen. Zum anderen sollen ausgewählte Regeln und  Empfehlungen für Leichte Sprache experimentell überprüft werden. Die  Ergebnisse werden mit dem Kooperationspartner und weiteren  Vertreter:innen der Zielgruppen diskutiert.


Doctoral Researcher/Postdoctoral Researcher (TV-L 13, 75%) – Der erfolgreiche Bewerber (w/m/d)  sollte einen Master in Psycholinguistik, Computerlinguistik,  theoretischer Linguistik oder in der Sprachwissenschaft des Deutschen  haben, ausreichend Erfahrung in experimenteller Arbeit nachweisen können  und das Deutsche auf muttersprachlichem Niveau beherrschen. Die  Aufgaben im Projekt beinhalten die Vorbereitung, Durchführung und  statistische Auswertung der geplanten Experimente. Der erfolgreiche  Bewerber (w/m/d) sollte mit modernen statistischen Methoden  (insbesondere mit Mixed Effects Models) vertraut sein. Die Arbeit im  Projekt gibt Gelegenheit zur Weiterqualifikation (Promotion).  Vertrautheit mit informationstheoretischen Konzepten und  computerlinguistischen Methoden ist erwünscht.