---
title: "Saarbruecken — research institutions and their papers"
description: "Saarbruecken (Germany): 4 institutions with papers on Elamur Research — 1 university, 3 labs and research institutes — and 58,971 citations of their works."
canonical: "https://research.elamur.ai/free/en/agglomerations/saarbruecken/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/agglomerations/saarbruecken/)

[home](https://md.research.elamur.ai/free/en/) / [institutions](https://md.research.elamur.ai/free/en/orgs/) / [agglomerations](https://md.research.elamur.ai/free/en/agglomerations/) / Saarbruecken

# Saarbruecken

Agglomeration [Germany](https://md.research.elamur.ai/free/en/countries/germany/) · metro population 570,240 (functional urban area, GHSL 2015)

**4** institutions with a page**96** works in our library**58,971** citations of their works

## About

### Profile

Saarbruecken is a city region in [Germany](https://md.research.elamur.ai/free/en/countries/germany/). On this site its research comes from four institutions: one university and three research institutes. [Saarland University](https://md.research.elamur.ai/free/en/universities/saarland-university/) contributes by far the largest share of the region's works here. The [Max Planck Institute for Informatics](https://md.research.elamur.ai/free/en/labs/max-planck-institute-for-informatics/), the [Helmholtz Center for Information Security](https://md.research.elamur.ai/free/en/labs/helmholtz-center-for-information-security/) and the [Max Planck Institute for Software Systems](https://md.research.elamur.ai/free/en/labs/max-planck-institute-for-software-systems/) add smaller bodies of work, each with its own focus.

Among the works in our library, the region's research is concentrated in computing, and above all in language technology. Most of it falls into a few areas: computational linguistics, the evaluation of machine translation, knowledge bases built from the web, the detection of harmful online content, and the safety and evaluation of large language models. One major contribution lies outside language. It is a reference paper on a core library for scientific computing in Python.

The institutions differ in how many works they have here and in how often those works are cited. Saarland University has the broadest collection. The Max Planck Institute for Software Systems has very few works here, but they are among the most cited of any institution in the region. The Helmholtz Center for Information Security is the most recent arrival in the collection and works on the security of language-model systems. The Max Planck Institute for Informatics sits between these in size. Its works cover knowledge harvesting, vision and language, and applied natural language processing.

### History

The earliest work from Saarbruecken on this site is [TnT – A Statistical Part-of-Speech Tagger](https://md.research.elamur.ai/free/en/papers/brants-2000-tnt-statistical-part-speech-tagger/) from 2000. It shows a statistical, corpus-driven style of computational linguistics at Saarland University. The paper takes a second-order hidden Markov model, tunes it with care, and shows that it assigns word classes as accurately as more elaborate methods.

The following decade continued in that vein, moving from word classes towards meaning. [Using Semantic Roles to Improve Question Answering](https://md.research.elamur.ai/free/en/papers/shen-2007-semantic-roles-improve-question-answering/) appeared in 2007. In 2009 came [The CoNLL-2009 Shared Task](https://md.research.elamur.ai/free/en/papers/hajic-2009-conll-2009-shared-task-syntactic/), a multilingual benchmark that joins syntactic dependency parsing with semantic role labelling.

From 2012 the Max Planck Institute for Informatics appears with knowledge-harvesting work. PATTY came in 2012, and in 2015 both YAGO3 and a study of cross-document co-reference followed. The middle of the 2010s also brought the region into large community evaluations of machine translation: the WMT findings reports of 2016 and 2019 and the 2017 biomedical translation task. In 2016 and 2017 the work spread further, to multimodal question answering, to hate speech detection and to mental-health signals in online forums.

Towards the end of the decade the range widened again. A reading-comprehension challenge set appeared in 2018, along with a system for checking the credibility of claims. The SciPy reference paper followed in 2019. In 2020 came an argument about whether language models can learn meaning from form alone. In 2021 the MasakhaNER work extended named entity recognition to African languages. From 2023 the Helmholtz Center for Information Security and the Max Planck Institute for Software Systems contribute papers on jailbreak prompts, prompt injection, trustworthiness and hard benchmarks for large language models. The latest of these, Humanity's Last Exam, dates from 2025.

### Research profile

Computational linguistics and core language processing form the oldest and deepest line in the region. TnT stands for the statistical tagging tradition. The CoNLL-2009 shared task overview and the 2007 paper on semantic roles for question answering show interest in predicate–argument structure. The 2007 paper builds a semantic structure for each question and for each candidate answer, pairing a predicate with weighted role assignments. It then compares these structures to pick out answers. [Climbing towards NLU](https://md.research.elamur.ai/free/en/papers/bender-2020-climbing-nlu-meaning-form-understanding/) adds a theoretical side. It counts as form every perceptible trace that language leaves, from text and pixels to bytes and the motions of the speech organs, and asks what a system could learn from such signals alone.

Machine translation evaluation is a second strong line. [Findings of the 2016 Conference on Machine Translation](https://md.research.elamur.ai/free/en/papers/bojar-2016-findings-2016-conference-machine-translation/) reports on the first WMT run as a full conference. [Findings of the 2019 Conference on Machine Translation (WMT19)](https://md.research.elamur.ai/free/en/papers/barrault-2019-findings-2019-conference-machine-translation/) continues that annual comparison of translation systems. Through the Max Planck Institute for Informatics the region also appears in [Findings of the WMT 2017 Biomedical Translation Shared Task](https://md.research.elamur.ai/free/en/papers/yepes-2017-findings-wmt-2017-biomedical-translation/), the second edition of a task that tests translation systems on biomedical text.

A third line is knowledge harvesting and knowledge bases, centred on the Max Planck Institute for Informatics. [YAGO3: A Knowledge Base from Multilingual Wikipedias](https://md.research.elamur.ai/free/en/papers/mahdisoltani-2015-yago3-knowledge-base-multilingual-wikipedias/) starts from the observation that knowledge bases drawn from Wikipedia tend to depend on its English edition. It builds instead on several language editions. [PATTY: A Taxonomy of Relational Patterns with Semantic Types](https://md.research.elamur.ai/free/en/papers/nakashole-2012-patty-taxonomy-relational-patterns-semantic/) tackles a related gap. Large knowledge resources name each relation with a bare label and miss the many ways texts express it. PATTY gathers such phrasings and arranges them into a taxonomy of typed patterns. [Cross-Document Co-Reference Resolution using Sample-Based Clustering with Knowledge](https://md.research.elamur.ai/free/en/papers/dutta-2015-cross-document-co-reference-resolution/) proposes CROCS, an unsupervised pipeline that links mentions of the same entity across documents and draws on knowledge-base information.

Online content, credibility and well-being form a fourth area. [A Survey on Hate Speech Detection using Natural Language Processing](https://md.research.elamur.ai/free/en/papers/schmidt-2017-survey-hate-speech-detection-natural/), from Saarland University, argues that filtering by keywords falls short because hatefulness depends on context. [DeClarE](https://md.research.elamur.ai/free/en/papers/popat-2018-declare-debunking-fake-news-false/) judges how credible a free-text claim is, for instance a rumour or a political statement, using evidence found on the web. [Depression and Self-Harm Risk Assessment in Online Forums](https://md.research.elamur.ai/free/en/papers/yates-2017-depression-self-harm-risk-assessment/) builds RSDD, a large corpus of Reddit users who report a depression diagnosis, and uses it to study risk signals in forum posts.

Reading comprehension and multilingual resources form a fifth area. [Looking Beyond the Surface](https://md.research.elamur.ai/free/en/papers/khashabi-2018-looking-beyond-surface-challenge-set/) presents MultiRC, a set of verified questions over paragraphs from several domains. Answering them requires combining information from more than one sentence. [MasakhaNER: Named Entity Recognition for African Languages](https://md.research.elamur.ai/free/en/papers/adelani-2021-masakhaner-named-entity-recognition-african/) builds named entity data for African languages. It also documents each language's typology and writing conventions, such as scripts and tone.

Vision and language appear through [Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding](https://md.research.elamur.ai/free/en/papers/fukui-2016-multimodal-compact-bilinear-pooling-visual/). The paper approximates the outer product of image and text features in compact form. This yields richer joint representations, both for answering questions about images and for locating phrases within them.

The newest line is the safety, security and evaluation of large language models. ["Do Anything Now"](https://md.research.elamur.ai/free/en/papers/shen-2023-anything-now-characterizing-evaluating-wild/) introduces JailbreakHub, which gathers prompts people use to get around model safeguards, studies them and tries them against models. [Not what you've signed up for](https://md.research.elamur.ai/free/en/papers/greshake-2023-ve-signed-up-compromising-real/) studies attacks on applications built around language models. It classifies how such attacks reach a system, whom they harm and what they achieve. [TrustLLM: Trustworthiness in Large Language Models](https://md.research.elamur.ai/free/en/papers/huang-2024-trustllm-trustworthiness-large-language-models/) responds to the absence of a shared framework for assessing how dependably these models act. [Humanity's Last Exam](https://md.research.elamur.ai/free/en/papers/phan-2025-humanity-s-last-exam/) answers the near-saturation of benchmarks like MMLU with a much harder test of model capability.

Scientific software forms a separate strand. [SciPy 1.0 — Fundamental Algorithms for Scientific Computing in Python](https://md.research.elamur.ai/free/en/papers/virtanen-2019-scipy-1-0-fundamental-algorithms/) presents an open-source package that has underpinned numerical work in Python since its earliest versions.

### Leading institutions and labs

Saarland University is the centre of the region's research on this site. Its works span the full arc from statistical tagging in 2000 to language-model research. They include the WMT findings of 2016 and 2019, the CoNLL-2009 overview, the hate speech survey, the MultiRC challenge set and the argument about meaning and form. Its profile is that of a broad computational linguistics and language technology community. It combines core methods, shared tasks, resources and conceptual debate.

The Max Planck Institute for Informatics covers knowledge harvesting with YAGO3, PATTY and CROCS. It also works on vision and language with compact bilinear pooling, on credibility with DeClarE, and on mental-health signals with the RSDD corpus. Multilingual and biomedical resources come through MasakhaNER and the 2017 biomedical translation task. Its works connect large structured knowledge to text understanding and to applied problems of the web.

The Helmholtz Center for Information Security brings an information-security view to language technology. All of its works here date from 2023 onward. They deal with jailbreaking, prompt injection, trustworthiness and capability benchmarks for large language models.

The Max Planck Institute for Software Systems has the smallest set of works here. It nonetheless includes two highly cited contributions, the SciPy 1.0 paper and Humanity's Last Exam. Both are large collective efforts, one in scientific software and one in model evaluation.

### Leading scientists

At Saarland University the first name is [Thorsten Brants](https://md.research.elamur.ai/free/en/authors/thorsten-brants/). Brants's TnT tagger is the earliest work from the region here and among the most cited, and it stands for an engineering-minded statistical approach to tagging. [Marcos Zampieri](https://md.research.elamur.ai/free/en/authors/marcos-zampieri/) follows, within the university's language technology community, where work on offensive and hateful content is one of the visible themes. [Vera Demberg](https://md.research.elamur.ai/free/en/authors/vera-demberg/) is associated with both Saarland University and the Max Planck Institute for Informatics, linking the two institutions. [Ernie Chang](https://md.research.elamur.ai/free/en/authors/ernie-chang/) is also among the university's people. So is [Jesujoba Alabi](https://md.research.elamur.ai/free/en/authors/jesujoba-alabi/), who likewise appears at the Max Planck Institute for Informatics, where the MasakhaNER resources for African languages are among the works. The university's list continues with [Wei Shi](https://md.research.elamur.ai/free/en/authors/wei-shi/), [Michael Wiegand](https://md.research.elamur.ai/free/en/authors/michael-wiegand/) and [Sebastian Schuster](https://md.research.elamur.ai/free/en/authors/sebastian-schuster/), all part of its natural language processing research.

At the Max Planck Institute for Informatics the first name is [Anna Rohrbach](https://md.research.elamur.ai/free/en/authors/anna-rohrbach/), followed by Jesujoba Alabi and [Gerhard Weikum](https://md.research.elamur.ai/free/en/authors/gerhard-weikum/). Weikum belongs to the institute where the knowledge-harvesting works were produced: YAGO3, PATTY, CROCS and DeClarE. [Andrew Yates](https://md.research.elamur.ai/free/en/authors/andrew-yates/) comes next and is the lead name on the study of depression and self-harm risk in online forums. [Fabian Suchanek](https://md.research.elamur.ai/free/en/authors/fabian-suchanek/) is also part of the institute's knowledge-base community. [Marcus Rohrbach](https://md.research.elamur.ai/free/en/authors/marcus-rohrbach/) appears in the institute's vision-and-language context, which includes the compact bilinear pooling work. [Gerard de Melo](https://md.research.elamur.ai/free/en/authors/gerard-de-melo/) and Vera Demberg complete the institute's first names.

At the Helmholtz Center for Information Security the first names are [Ziqing Yang](https://md.research.elamur.ai/free/en/authors/ziqing-yang/), [Yang Zhang](https://md.research.elamur.ai/free/en/authors/yang-zhang/) and [Martin Jaggi](https://md.research.elamur.ai/free/en/authors/martin-jaggi/). Their institution's works here concern jailbreak prompts, prompt injection and the trustworthiness of language models.

At the Max Planck Institute for Software Systems the names are [Jan Leike](https://md.research.elamur.ai/free/en/authors/jan-leike/), [Abhilasha Ravichander](https://md.research.elamur.ai/free/en/authors/abhilasha-ravichander/) and [Zhiqiang Xie](https://md.research.elamur.ai/free/en/authors/zhiqiang-xie/). Their institute's works here cover scientific computing in Python and the evaluation of frontier language models.

### Landmark works

The most cited work from the region on this site is TnT – A Statistical Part-of-Speech Tagger. It showed that a trigram hidden Markov model, tuned with care, could match more complex taggers.

Close behind is the hate speech survey from Saarland University. It reviews the features and methods used for detecting hateful messages and stresses that context matters more than word lists. The 2016 WMT findings rank next. They report the results of the shared tasks in the first edition run as a full conference.

The SciPy 1.0 paper, credited to the Max Planck Institute for Software Systems, is among the most cited works in the region. It documents the algorithms, design and community development of a library basic to numerical computing. The 2019 WMT findings continue the annual reporting of translation results.

Climbing towards NLU is one of the most cited conceptual works here. It argues that a system trained only on form has no route to meaning, and it offers precise terms for discussing what large language models can and cannot grasp. MultiRC and the CoNLL-2009 overview are both much cited benchmark papers. The first covers multi-sentence reading comprehension, the second multilingual parsing joined with semantic role labelling.

Among the newer works, "Do Anything Now" is the most cited of the security papers. It is followed by Humanity's Last Exam, Not what you've signed up for and TrustLLM. Together they show how quickly the region's language-model safety research has gained attention on this site. At the Max Planck Institute for Informatics, the most cited works are MasakhaNER, YAGO3, the compact bilinear pooling paper and PATTY.

### Collaborations

The clearest links between institutions in the region run through shared people and shared works. Vera Demberg and Jesujoba Alabi each appear at both Saarland University and the Max Planck Institute for Informatics. Language research therefore moves between the university and the institute in both directions.

Humanity's Last Exam is credited to both the Helmholtz Center for Information Security and the Max Planck Institute for Software Systems. It ties together the region's two newest strands of language-model research.

Many of the region's most cited works are large community efforts rather than single-lab papers. These include the WMT findings, the CoNLL-2009 shared task, the biomedical translation task, MasakhaNER, SciPy 1.0, TrustLLM and Humanity's Last Exam. Through them, Saarbruecken researchers take part in wide international teams that build shared benchmarks, datasets and software. The region's institutions on this site are all academic and public research bodies, and no company appears among them. Links between industry and academia are therefore not visible here.

### Timeline

In 2000 the region's work in our library begins with the TnT tagger at Saarland University. Statistical computational linguistics is the founding line.

The years 2007 and 2009 extend that line from word classes to meaning. Semantic roles are applied to question answering, and the CoNLL-2009 shared task joins syntactic and semantic dependencies across languages.

From 2012 to 2015 the Max Planck Institute for Informatics contributes knowledge-harvesting work. PATTY appears in 2012, and YAGO3 and the CROCS co-reference pipeline follow in 2015.

The years 2016 to 2018 bring the widest spread of topics. In 2016 the region takes part in the WMT findings and introduces compact bilinear pooling for visual question answering. In 2017 it contributes the hate speech survey, the RSDD depression corpus and the biomedical translation findings. MultiRC and DeClarE follow in 2018.

From 2019 to 2021 come the WMT19 findings and the SciPy 1.0 paper in 2019, the meaning-and-form argument in 2020, and MasakhaNER in 2021.

From 2023 to 2025 the focus turns to large language models and their risks. Jailbreak prompts and prompt injection are studied in 2023, TrustLLM appears in 2024, and Humanity's Last Exam in 2025. The Helmholtz Center for Information Security and the Max Planck Institute for Software Systems lead this most recent strand of the region's research on this site.

## Institutions (4)

1 university, 3 labs and research institutes.

### 🎓 Universities (1)

1. **[Saarland University](https://md.research.elamur.ai/free/en/universities/saarland-university/)**University · Saarbrücken, [Germany](https://md.research.elamur.ai/free/en/countries/germany/)*86 works · 15,761 citations*

### 🔬 Labs and research institutes (3)

1. **[Max Planck Institute for Informatics](https://md.research.elamur.ai/free/en/labs/max-planck-institute-for-informatics/)**Research institute · Saarbrücken, [Germany](https://md.research.elamur.ai/free/en/countries/germany/)*9 works · 3,657 citations*
2. **[Helmholtz Center for Information Security](https://md.research.elamur.ai/free/en/labs/helmholtz-center-for-information-security/)**Research institute · Saarbrücken, [Germany](https://md.research.elamur.ai/free/en/countries/germany/)*4 works · 1,001 citations*
3. **[Max Planck Institute for Software Systems](https://md.research.elamur.ai/free/en/labs/max-planck-institute-for-software-systems/)**Research institute · Saarbrücken, [Germany](https://md.research.elamur.ai/free/en/countries/germany/)*2 works · 38,552 citations*

## Works in our library (96)

Works of its institutions, the most cited first.

1. [**SciPy 1.0 — Fundamental Algorithms for Scientific Computing in Python**Pauli Virtanen, Ralf Gommers, Travis E. Oliphant, et al.*2019 · 38,317 citations*](https://md.research.elamur.ai/free/en/papers/virtanen-2019-scipy-1-0-fundamental-algorithms/)
2. [**A Survey on Hate Speech Detection using Natural Language Processing**Anna Schmidt, Michael Wiegand*2017 · 1,799 citations*](https://md.research.elamur.ai/free/en/papers/schmidt-2017-survey-hate-speech-detection-natural/)
3. [**TnT – A Statistical Part-of-Speech Tagger**Thorsten Brants*2000 · 1,756 citations*](https://md.research.elamur.ai/free/en/papers/brants-2000-tnt-statistical-part-speech-tagger/)
4. [**Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding**Akira Fukui, Dong Huk Park, Daylen Yang, et al.*2016 · 1,642 citations*](https://md.research.elamur.ai/free/en/papers/fukui-2016-multimodal-compact-bilinear-pooling-visual/)
5. [**Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data**Emily M. Bender, Alexander Koller*2020 · 1,371 citations*](https://md.research.elamur.ai/free/en/papers/bender-2020-climbing-nlu-meaning-form-understanding/)
6. [**Findings of the 2016 Conference on Machine Translation**Ondřej Bojar, Rajen Chatterjee, Christian Federmann, et al.*2016 · 936 citations*](https://md.research.elamur.ai/free/en/papers/bojar-2016-findings-2016-conference-machine-translation/)
7. [**Findings of the 2019 Conference on Machine Translation (WMT19)**Loïc Barrault, Ondřej Bojar, Marta R. Costa-jussà, et al.*2019 · 839 citations*](https://md.research.elamur.ai/free/en/papers/barrault-2019-findings-2019-conference-machine-translation/)
8. [**Looking Beyond the Surface: A Challenge Set for Reading Comprehension over Multiple Sentences**Daniel Khashabi, Snigdha Chaturvedi, Michael Roth, et al.*2018 · 669 citations*](https://md.research.elamur.ai/free/en/papers/khashabi-2018-looking-beyond-surface-challenge-set/)
9. [**The CoNLL-2009 Shared Task: Syntactic and Semantic Dependencies in Multiple Languages**Jan Hajič, Massimiliano Ciaramita, Richard Johansson, et al.*2009 · 631 citations*](https://md.research.elamur.ai/free/en/papers/hajic-2009-conll-2009-shared-task-syntactic/)
10. [**Using Semantic Roles to Improve Question Answering**Dan Shen, Mirella Lapata*2007 · 522 citations*](https://md.research.elamur.ai/free/en/papers/shen-2007-semantic-roles-improve-question-answering/)
11. [**Inducing Crosslingual Distributed Representations of Words**Alexandre Klementiev, Ivan Titov, Binod Bhattarai*2012 · 501 citations*](https://md.research.elamur.ai/free/en/papers/klementiev-2012-inducing-crosslingual-distributed-representations-words/)
12. [**PATTY: A Taxonomy of Relational Patterns with Semantic Types**Ndapandula Nakashole, Gerhard Weikum, Fabian Suchanek*2012 · 432 citations*](https://md.research.elamur.ai/free/en/papers/nakashole-2012-patty-taxonomy-relational-patterns-semantic/)
13. [**MasakhaNER: Named Entity Recognition for African Languages**David Ifeoluwa Adelani, Jade Abbott, Graham Neubig, et al.*2021 · 429 citations*](https://md.research.elamur.ai/free/en/papers/adelani-2021-masakhaner-named-entity-recognition-african/)
14. [**75 Languages, 1 Model: Parsing Universal Dependencies Universally**Dan Kondratyuk, Milan Straka*2019 · 415 citations*](https://md.research.elamur.ai/free/en/papers/kondratyuk-2019-75-languages-1-model-parsing/)
15. [**An Annotation Scheme for Free Word Order Languages**Wojciech Skut, Brigitte Krenn, Thorsten Brants, et al.*1997 · 408 citations*](https://md.research.elamur.ai/free/en/papers/skut-1997-annotation-scheme-free-word-order/)
16. [**Findings of the 2020 Conference on Machine Translation (WMT20)**Loïc Barrault, Magdalena Biesialska, Ondřej Bojar, et al.*2020 · 401 citations*](https://md.research.elamur.ai/free/en/papers/barrault-2020-findings-2020-conference-machine-translation/)
17. [**Depression and Self-Harm Risk Assessment in Online Forums**Andrew Yates, Arman Cohan, Nazli Goharian*2017 · 401 citations*](https://md.research.elamur.ai/free/en/papers/yates-2017-depression-self-harm-risk-assessment/)
18. [**A Survey on Recent Approaches for Natural Language Processing in Low-Resource Scenarios**Michael A. Hedderich, Lukas Lange, Heike Adel, et al.*2021 · 399 citations*](https://md.research.elamur.ai/free/en/papers/hedderich-2021-survey-recent-approaches-natural-language/)
19. [**DeClarE: Debunking Fake News and False Claims using Evidence-Aware Deep Learning**Kashyap Popat, Subhabrata Mukherjee, Andrew Yates, et al.*2018 · 388 citations*](https://md.research.elamur.ai/free/en/papers/popat-2018-declare-debunking-fake-news-false/)
20. [**Participatory Research for Low-resourced Machine Translation: A Case Study in African Languages**Wilhelmina Nekoto, Vukosi Marivate, Tshinondiwa Matsila, et al.*2020 · 333 citations*](https://md.research.elamur.ai/free/en/papers/nekoto-2020-participatory-research-low-resourced-machine/)
21. [**"Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models**Xinyue Shen, Zeyuan Chen, Michael Backes, et al.*2023 · 320 citations*](https://md.research.elamur.ai/free/en/papers/shen-2023-anything-now-characterizing-evaluating-wild/)
22. [**Experiments on the Automatic Induction of German Semantic Verb Classes**Schulte im Walde, Sabine*2006 · 259 citations*](https://md.research.elamur.ai/free/en/papers/walde-2006-experiments-automatic-induction-german-semantic/)
23. [**The SALSA Corpus: a German Corpus Resource for Lexical Semantics**Aljoscha Burchardt, Katrin Erk, Anette Frank, et al.*2006 · 255 citations*](https://md.research.elamur.ai/free/en/papers/burchardt-2006-salsa-corpus-german-corpus-resource/)
24. [**YAGO3: A Knowledge Base from Multilingual Wikipedias**Farzaneh Mahdisoltani, Joanna Biega, Fabian M. Suchanek*2015 · 235 citations*](https://md.research.elamur.ai/free/en/papers/mahdisoltani-2015-yago3-knowledge-base-multilingual-wikipedias/)
25. [**Humanity's Last Exam**Long Phan, Alice Gatti, Ziwen Han, et al.*2025 · 235 citations*](https://md.research.elamur.ai/free/en/papers/phan-2025-humanity-s-last-exam/)

[See all 96 works →](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/works/)

## Works by year

### 2020s

[**2026** 7 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2026/) · [**2025** 4 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2025/) · [**2024** 3 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2024/) · [**2023** 10 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2023/) · [**2022** 7 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2022/) · [**2021** 3 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2021/) · [**2020** 5 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2020/)

### 2010s

[**2019** 7 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2019/) · [**2018** 5 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2018/) · [**2017** 9 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2017/) · [**2016** 8 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2016/) · [**2015** 3 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2015/) · [**2014** 1 work](https://md.research.elamur.ai/free/en/papers/zampieri-2014-report-dsl-shared-task/) · [**2012** 6 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2012/) · [**2011** 1 work](https://md.research.elamur.ai/free/en/papers/bender-2011-parser-evaluation-over-local-non/) · [**2010** 1 work](https://md.research.elamur.ai/free/en/papers/palmer-2010-evaluating-framenet-style-semantic-parsing/)

### 2000s

[**2009** 2 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2009/) · [**2008** 4 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2008/) · [**2007** 1 work](https://md.research.elamur.ai/free/en/papers/shen-2007-semantic-roles-improve-question-answering/) · [**2006** 4 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2006/) · [**2004** 2 works](https://md.research.elamur.ai/free/en/agglomerations/saarbruecken/year/2004/) · [**2000** 1 work](https://md.research.elamur.ai/free/en/papers/brants-2000-tnt-statistical-part-speech-tagger/)

### 1990s

[**1997** 1 work](https://md.research.elamur.ai/free/en/papers/skut-1997-annotation-scheme-free-word-order/) · [**1994** 1 work](https://md.research.elamur.ai/free/en/papers/bos-1994-presupposition-vp-ellipsis/)

## Concepts, shelves and topics

Of its works in our library: the number of works on each and its rating among them.

💡 Concepts 8[Part-of-speech tagging · 1 work](https://md.research.elamur.ai/free/en/concepts/part-of-speech-tagging/) · [Semantic role labeling · 1 work](https://md.research.elamur.ai/free/en/concepts/semantic-role-labeling/) · [Named entity recognition · 1 work](https://md.research.elamur.ai/free/en/concepts/named-entity-recognition/) · [Universal Dependencies · 1 work](https://md.research.elamur.ai/free/en/concepts/universal-dependencies/) · [Lexical semantics · 1 work](https://md.research.elamur.ai/free/en/concepts/lexical-semantics/) · [Knowledge representation and reasoning · 1 work](https://md.research.elamur.ai/free/en/concepts/knowledge-representation/) · [Corpus linguistics · 1 work](https://md.research.elamur.ai/free/en/concepts/corpus-linguistics/) · [Implicature · 1 work](https://md.research.elamur.ai/free/en/concepts/implicature/)

📚 Shelves 11[Natural language processing · 21 works](https://md.research.elamur.ai/free/en/shelves/nlp/) · [Linguistics & languages · 66 works](https://md.research.elamur.ai/free/en/shelves/linguistics/) · [Mathematics & statistics · 2 works](https://md.research.elamur.ai/free/en/shelves/mathematics/) · [Other · 5 works](https://md.research.elamur.ai/free/en/shelves/other/) · [English · 18 works](https://md.research.elamur.ai/free/en/shelves/en/) · [Machine translation · 11 works](https://md.research.elamur.ai/free/en/shelves/machine-translation/) · [AI & machine learning · 3 works](https://md.research.elamur.ai/free/en/shelves/ai/) · [Law · 1 work](https://md.research.elamur.ai/free/en/shelves/law/) · [Computing & GPUs · 2 works](https://md.research.elamur.ai/free/en/shelves/computing/) · [Machine translation · 4 works](https://md.research.elamur.ai/free/en/shelves/mtrans/) · [History & humanities · 1 work](https://md.research.elamur.ai/free/en/shelves/history/)

🧭 Topics 35[Discourse analysis · 18 works](https://md.research.elamur.ai/free/en/topics/discourse-analysis/) · [Multilingual & cross-lingual NLP · 11 works](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/) · [Evaluation & benchmarks · 11 works](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · [Machine translation · 8 works](https://md.research.elamur.ai/free/en/topics/machine-translation/) · [Question answering · 4 works](https://md.research.elamur.ai/free/en/topics/question-answering/) · [Semantic role labeling · 6 works](https://md.research.elamur.ai/free/en/topics/semantic-role-labeling/) · [Part-of-speech tagging · 1 work](https://md.research.elamur.ai/free/en/topics/part-of-speech-tagging/) · [Language models & pretraining · 5 works](https://md.research.elamur.ai/free/en/topics/language-models/) · [Bias, fairness & safety · 3 works](https://md.research.elamur.ai/free/en/topics/bias-fairness-safety/) · [Named entity recognition · 2 works](https://md.research.elamur.ai/free/en/topics/named-entity-recognition/) · [Low-resource languages · 3 works](https://md.research.elamur.ai/free/en/topics/low-resource-languages/) · [Coreference resolution · 7 works](https://md.research.elamur.ai/free/en/topics/coreference-resolution/) · [Sentiment & emotion · 2 works](https://md.research.elamur.ai/free/en/topics/sentiment-emotion/) · [Universal Dependencies · 4 works](https://md.research.elamur.ai/free/en/topics/universal-dependencies/) · [Vision & language · 1 work](https://md.research.elamur.ai/free/en/topics/multimodal/)+20 more

Agglomerations: the GHSL Urban Centre Database 2025; populations: the GHSL functional urban areas 2015 (urban centre and commuting zone), else the urban centre, as each page says (European Commission, Joint Research Centre, CC BY 4.0). An institution belongs to the urban centre its coordinates (ROR, OpenAlex) fall in, or one within 5 km. Land: Natural Earth.

- [← back to](https://md.research.elamur.ai/free/en/countries/germany/) — Germany

Updated 23 October 2026