---
title: "Switzerland — research institutions, scientists and papers"
description: "Switzerland: 10 institutions with papers in our library in 7 agglomerations, 71 people with a page, 123 works, 52,398 citations."
canonical: "https://research.elamur.ai/free/en/countries/switzerland/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/countries/switzerland/)

[home](https://md.research.elamur.ai/free/en/) / [countries](https://md.research.elamur.ai/free/en/countries/) / Switzerland

# Switzerland

Country 10 institutions in 7 agglomerations · [all countries](https://md.research.elamur.ai/free/en/countries/)

[**10** institutions with a page](https://md.research.elamur.ai/free/en/countries/switzerland/#institutions) · [**7** agglomerations](https://md.research.elamur.ai/free/en/countries/switzerland/#agglomerations) · [**71** people with a page](https://md.research.elamur.ai/free/en/countries/switzerland/#authors) · [**123** works in our library](https://md.research.elamur.ai/free/en/countries/switzerland/#works) · [**52,398** citations of their works](https://md.research.elamur.ai/free/en/countries/switzerland/#works)

## About

### Profile

Switzerland appears on this site through a small and varied group of research institutions spread across several cities and more than one of the country's language regions. Most of them are universities. Alongside them are one independent research institute, the [Idiap Research Institute](https://md.research.elamur.ai/free/en/labs/idiap-research-institute/) in Martigny-Combe in the Valais, and one company office, [Google (Switzerland)](https://md.research.elamur.ai/free/en/companies/google-switzerland/) in Zurich.

The universities include the two federal institutes of technology, [ETH Zurich](https://md.research.elamur.ai/free/en/universities/eth-zurich/) and the [École Polytechnique Fédérale de Lausanne](https://md.research.elamur.ai/free/en/universities/ecole-polytechnique-federale-de-lausanne/). There are also the [University of Zurich](https://md.research.elamur.ai/free/en/universities/university-of-zurich/), the [University of Geneva](https://md.research.elamur.ai/free/en/universities/university-of-geneva/), the [University of Bern](https://md.research.elamur.ai/free/en/universities/university-of-bern/), the [University of Fribourg](https://md.research.elamur.ai/free/en/universities/university-of-fribourg/), the [University of St.Gallen](https://md.research.elamur.ai/free/en/universities/university-of-st-gallen/) and the [Università della Svizzera italiana](https://md.research.elamur.ai/free/en/universities/universita-della-svizzera-italiana/) in Lugano.

[Zurich](https://md.research.elamur.ai/free/en/agglomerations/zurich/) is the clearest centre. It hosts ETH Zurich, the University of Zurich and the Google office, and much of the Swiss work in our library comes from that city. A second cluster sits around Lake Geneva, with [Lausanne](https://md.research.elamur.ai/free/en/agglomerations/lausanne/) and [Geneva](https://md.research.elamur.ai/free/en/agglomerations/geneva/), and the Idiap institute lies close to it in the Valais. Smaller contributions come from [Lugano](https://md.research.elamur.ai/free/en/agglomerations/lugano/), [Bern](https://md.research.elamur.ai/free/en/agglomerations/bern/), [Freiburg im Üechtland](https://md.research.elamur.ai/free/en/agglomerations/freiburg-im-uechtland/) and [Sankt Gallen](https://md.research.elamur.ai/free/en/agglomerations/st-gallen/).

The Swiss works here are concentrated in language research. They cover linguistics, natural language processing and machine translation, with smaller strands in machine learning, audio, law and mathematics. Everything below describes the works and people on this site only. It is not an account of Swiss research as a whole.

### History

The earliest Swiss works here come from Geneva and concern how language technology should be judged and built. A 1990 paper by King, [Using Test Suites in Evaluation of Machine Translation Systems](https://md.research.elamur.ai/free/en/papers/king-1990-test-suites-evaluation-machine-translation/), holds that no single evaluation suits every buyer of a translation system. It proposes two kinds of test suite instead. Five years later, [Identifying Word Translations in Non-Parallel Texts](https://md.research.elamur.ai/free/en/papers/rapp-1995-identifying-word-translations-non-parallel/) by Reinhard Rapp tried to find translation equivalents without aligned bilingual text. It began with English stimulus words from a classic association list and their German counterparts. Geneva's line continued in 2003 with Alexander Clark's work on part-of-speech induction and in 2006 with Paola Merlo's study of prepositional phrase attachment.

From around 2010 a discourse-oriented strand appears, centred on the Idiap Research Institute and on Geneva. It includes dialogue-act annotation standards in 2010 and 2012, discourse connectives in statistical machine translation in 2012 and 2013, and annotation of connective meaning in 2013. Corporate research from Zurich enters in 2011 with work on cross-lingual dependency parsing. Fribourg adds discourse-relation studies in 2015.

The middle of the 2010s marks a turn towards neural methods. In 2016 Lausanne produced work on graph neural networks and on neural text generation. In 2017 ETH Zurich contributed character-level neural translation. The years from 2018 onward are the densest. Zurich, Geneva, Lausanne and Idiap all added neural machine translation, embeddings and analyses of what neural networks learn. After 2020 the emphasis moves again, towards large language models, multimodal pretraining, audio generation, legal benchmarks and safety. The most recent works here date from 2025.

### Research profile

Machine translation is the largest topic among the Swiss works, and evaluation and benchmarks come second. Much of the translation work asks how systems handle text beyond the single sentence. [Context-Aware Neural Machine Translation Learns Anaphora Resolution](https://md.research.elamur.ai/free/en/papers/voita-2018-context-aware-neural-machine-translation/) shows that a model given surrounding sentences learns to resolve pronouns. [When a Good Translation is Wrong in Context](https://md.research.elamur.ai/free/en/papers/voita-2019-when-good-translation-is-wrong/) looks at deixis, ellipsis and lexical cohesion. From Idiap, [Document-Level Neural Machine Translation with Hierarchical Attention Networks](https://md.research.elamur.ai/free/en/papers/miculicich-2018-document-level-neural-machine-translation/) moves past the standard practice of translating each sentence on its own. On the evaluation side, [Has Machine Translation Achieved Human Parity?](https://md.research.elamur.ai/free/en/papers/laubli-2018-has-machine-translation-achieved-human/) revisits a claim that a neural Chinese–English system matched professional translators. It argues that judging whole documents gives a different picture. Idiap's metric for pronoun translation accuracy, APT, belongs to the same concern with measuring translation quality. Multilingual work includes [Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation](https://md.research.elamur.ai/free/en/papers/zhang-2020-improving-massively-multilingual-neural-machine/), which introduces random online backtranslation, and the overview reports of the WMT19 and WMT24 shared tasks.

[Discourse analysis](https://md.research.elamur.ai/free/en/topics/discourse-analysis/) forms a distinctive Swiss strand. Idiap studied how ambiguous English connectives can be sense-labelled to improve statistical translation. It also examined how often an explicit English connective has no counterpart in a translation, a phenomenon called implicitation. Geneva proposed translation spotting as a way to annotate connective meaning. Fribourg contributed [Factors Influencing the Implicitation of Discourse Relations across Languages](https://md.research.elamur.ai/free/en/papers/hoek-2015-factors-influencing-implicitation-discourse-relations/), which proposes that the relations most easily left implicit are the cognitively simplest. A companion study by Crible and Zufferey tests whether one sense taxonomy can annotate discourse markers in both speech and writing. The ISO 24617-2 dialogue-act standard, described at a draft stage in 2010 and in its approved form in 2012, links this strand to [dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/).

A third line asks what neural models learn about linguistic structure. [Colorless Green Recurrent Networks Dream Hierarchically](https://md.research.elamur.ai/free/en/papers/gulordava-2018-colorless-green-recurrent-networks-dream/) from Geneva tests whether networks trained only to predict the next word acquire abstract syntax. [Information-Theoretic Probing for Linguistic Structure](https://md.research.elamur.ai/free/en/papers/pimentel-2020-information-theoretic-probing-linguistic-structure/) from ETH Zurich recasts probing in information-theoretic terms, using control functions that strip away context as baselines. [Analyzing Multi-Head Self-Attention](https://md.research.elamur.ai/free/en/papers/voita-2019-analyzing-multi-head-self-attention/) finds that a few specialised attention heads carry most of the work in translation encoders. St.Gallen's studies of negation and of pragmatic categories in inference add a philosophical angle to the same question.

[Language models](https://md.research.elamur.ai/free/en/topics/language-models/) and [reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) define the recent period. ETH Zurich's [Distilling Reasoning Capabilities into Smaller Language Models](https://md.research.elamur.ai/free/en/papers/shridhar-2023-distilling-reasoning-capabilities-into-smaller/) transfers chain-of-thought ability to smaller models. Another ETH work argues that transformers learn in context by carrying out something like gradient descent. A third, on refusal, finds that refusal behaviour in chat models is mediated by a single direction in their activations, which ties it to [interpretability](https://md.research.elamur.ai/free/en/topics/interpretability/). Several institutions share [Humanity's Last Exam](https://md.research.elamur.ai/free/en/papers/phan-2025-humanity-s-last-exam/), a benchmark built because older tests such as MMLU had become nearly saturated.

Smaller strands show the range of the collection. Bern leads in [legal AI](https://md.research.elamur.ai/free/en/topics/legal-ai/) with Swiss-Judgment-Prediction and the LEXTREME benchmark. Google's Zurich office covers [speech](https://md.research.elamur.ai/free/en/topics/speech/) and audio with SoundStream and AudioLM, and [vision and language](https://md.research.elamur.ai/free/en/topics/multimodal/) with sigmoid-loss image–text pretraining. Lausanne's graph convolution work and the NumPy paper reach into mathematics and scientific computing.

### Leading scientists

The first people shown for Switzerland are [Caglar Gulcehre](https://md.research.elamur.ai/free/en/authors/caglar-gulcehre/), affiliated with the École Polytechnique Fédérale de Lausanne, and [Rico Sennrich](https://md.research.elamur.ai/free/en/authors/rico-sennrich/) of the University of Zurich. Sennrich has the largest body of work here of anyone in the country. It centres on machine translation in context: anaphora resolution in context-aware translation, deixis and ellipsis, and the human-parity question raised with Samuel Läubli and Martin Volk.

Three researchers at Google (Switzerland) come next. [Basil Mustafa](https://md.research.elamur.ai/free/en/authors/basil-mustafa/) is a co-author of the sigmoid-loss work on language–image pretraining. [Julian Eisenschlos](https://md.research.elamur.ai/free/en/authors/julian-eisenschlos/) is also affiliated with the Zurich office. [Juraj Gottweis](https://md.research.elamur.ai/free/en/authors/juraj-gottweis/) is the first author of [Accelerating scientific discovery with Co-Scientist](https://md.research.elamur.ai/free/en/papers/gottweis-2025-accelerating-scientific-discovery-co/), which presents a multi-agent system built on Gemini.

[Antoine Bosselut](https://md.research.elamur.ai/free/en/authors/antoine-bosselut/) works at Lausanne, whose output here runs from text generation and embeddings to entity linking and multi-task fine-tuning. [Florian Tramer](https://md.research.elamur.ai/free/en/authors/florian-tramer/) appears with both Google (Switzerland) and ETH Zurich, two of the Zurich institutions whose recent work covers language-model behaviour and safety. [Marco Tagliasacchi](https://md.research.elamur.ai/free/en/authors/marco-tagliasacchi/) belongs to the Google office, whose work includes the neural audio line.

[Ryan Cotterell](https://md.research.elamur.ai/free/en/authors/ryan-cotterell/) has a substantial set of works at ETH Zurich, where the institution's research includes probing and the analysis of linguistic structure in models. [Keith Hall](https://md.research.elamur.ai/free/en/authors/keith-hall/) of Google (Switzerland) is a co-author of [Multi-Source Transfer of Delexicalized Dependency Parsers](https://md.research.elamur.ai/free/en/papers/mcdonald-2011-multi-source-transfer-delexicalized-dependency/). [Damien Vincent](https://md.research.elamur.ai/free/en/authors/damien-vincent/), also at the Google office, co-wrote [AudioLM](https://md.research.elamur.ai/free/en/papers/borsos-2022-audiolm-language-modeling-approach-audio/). [Thomas Hofmann](https://md.research.elamur.ai/free/en/authors/thomas-hofmann/) of ETH Zurich co-wrote [Fully Character-Level Neural Machine Translation without Explicit Segmentation](https://md.research.elamur.ai/free/en/papers/lee-2017-fully-character-level-neural-machine/) with Jason Lee and Kyunghyun Cho.

### Landmark works

The most cited Swiss work here by a wide margin is [Array Programming with NumPy](https://md.research.elamur.ai/free/en/papers/harris-2020-array-programming-numpy/), attached to Google (Switzerland). It describes the central library for array programming in Python, so its reach extends well beyond language research.

Next comes [Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering](https://md.research.elamur.ai/free/en/papers/defferrard-2016-convolutional-neural-networks-graphs-fast/) by Michael Defferrard, [Xavier Bresson](https://md.research.elamur.ai/free/en/authors/xavier-bresson/) and Pierre Vandergheynst of Lausanne. It starts from the fact that classical convolutional networks rely on regular grids. It then defines convolution and pooling on graphs through fast localised spectral filters.

From the University of Zurich, [ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks](https://md.research.elamur.ai/free/en/papers/gilardi-2023-chatgpt-outperforms-crowd-workers-text/) by Fabrizio Gilardi, Meysam Alizadeh and Maël Kubli tested whether gpt-3.5-turbo, used zero-shot, could take the place of human crowd annotators. The analysis of multi-head self-attention follows closely among the most cited works.

[Sigmoid Loss for Language Image Pre-Training](https://md.research.elamur.ai/free/en/papers/zhai-2023-sigmoid-loss-language-image-pre/), led by [Xiaohua Zhai](https://md.research.elamur.ai/free/en/authors/xiaohua-zhai/), replaces the usual contrastive objective with a pairwise sigmoid loss. Also from 2023, the University of Zurich's [Adapted Large Language Models Can Outperform Medical Experts in Clinical Text Summarization](https://md.research.elamur.ai/free/en/papers/veen-2023-adapted-large-language-models-outperform/) addresses the time clinicians spend condensing health records into reports.

[Unsupervised Learning of Sentence Embeddings Using Compositional n-Gram Features](https://md.research.elamur.ai/free/en/papers/pagliardini-2018-unsupervised-learning-sentence-embeddings-compositional/), by [Matteo Pagliardini](https://md.research.elamur.ai/free/en/authors/matteo-pagliardini/), [Prakhar Gupta](https://md.research.elamur.ai/free/en/authors/prakhar-gupta/) and [Martin Jaggi](https://md.research.elamur.ai/free/en/authors/martin-jaggi/), introduced Sent2Vec. It extends the C-BOW objective from a local window to the whole sentence.

Other works widely cited on this site include the WMT19 findings, Gulordava's recurrent-network study, SoundStream, Lausanne's biography-generation paper and [Learning Word Vectors for 157 Languages](https://md.research.elamur.ai/free/en/papers/grave-2018-learning-word-vectors-157-languages/). The last of these answers the scarcity of pretrained embeddings outside English. [Scalable Zero-shot Entity Linking with Dense Entity Retrieval](https://md.research.elamur.ai/free/en/papers/wu-2020-scalable-zero-shot-entity-linking/) shows that BERT-based models can link entities never seen in training.

### Labs and units

The ten Swiss institutions differ greatly in what they bring. The University of Zurich has the broadest output here. It is anchored in machine translation and multilingual data, including the audit of web-crawled multilingual corpora [Quality at a Glance](https://md.research.elamur.ai/free/en/papers/kreutzer-2021-quality-glance-audit-web-crawled/). Its people include [Mathias Müller](https://md.research.elamur.ai/free/en/authors/mathias-muller/), [Annette Rios](https://md.research.elamur.ai/free/en/authors/annette-rios/), [Martin Volk](https://md.research.elamur.ai/free/en/authors/martin-volk/), [Samuel Läubli](https://md.research.elamur.ai/free/en/authors/samuel-laubli/), [Tanja Samardžić](https://md.research.elamur.ai/free/en/authors/tanja-samardzic/) and [Simon Clematide](https://md.research.elamur.ai/free/en/authors/simon-clematide/).

ETH Zurich concentrates on language models, probing, reasoning and machine translation evaluation. It also includes [Mrinmaya Sachan](https://md.research.elamur.ai/free/en/authors/mrinmaya-sachan/), [Kumar Shridhar](https://md.research.elamur.ai/free/en/authors/kumar-shridhar/), [Clara Meister](https://md.research.elamur.ai/free/en/authors/clara-meister/) and [Vilém Zouhar](https://md.research.elamur.ai/free/en/authors/vilem-zouhar/).

The École Polytechnique Fédérale de Lausanne combines machine learning on graphs with embeddings, generation and parameter-efficient fine-tuning. [Robert West](https://md.research.elamur.ai/free/en/authors/robert-west/) is among its people.

The University of Geneva carries the oldest lines here, in evaluation, translation lexicons and syntax. Its people include [Paola Merlo](https://md.research.elamur.ai/free/en/authors/paola-merlo/), [Kristina Gulordava](https://md.research.elamur.ai/free/en/authors/kristina-gulordava/), [Yves Scherrer](https://md.research.elamur.ai/free/en/authors/yves-scherrer/) and [James Henderson](https://md.research.elamur.ai/free/en/authors/james-henderson/). Geneva also took part in [the VarDial 2017 campaign](https://md.research.elamur.ai/free/en/papers/zampieri-2017-findings-vardial-evaluation-campaign/) on closely related languages and dialects.

The Idiap Research Institute works across discourse, dialogue standards, document-level translation and bias mitigation. Its people include [Rabeeh Karimi Mahabadi](https://md.research.elamur.ai/free/en/authors/rabeeh-karimi-mahabadi/), [Nikolaos Pappas](https://md.research.elamur.ai/free/en/authors/nikolaos-pappas/) and [Thomas Meyer](https://md.research.elamur.ai/free/en/authors/thomas-meyer/).

The smaller institutions each have a defined focus. Bern works on legal benchmarks with [Joel Niklaus](https://md.research.elamur.ai/free/en/authors/joel-niklaus/). St.Gallen covers pragmatics with [Reto Gubelmann](https://md.research.elamur.ai/free/en/authors/reto-gubelmann/) and [Siegfried Handschuh](https://md.research.elamur.ai/free/en/authors/siegfried-handschuh/). Fribourg studies discourse markers with [Sandrine Zufferey](https://md.research.elamur.ai/free/en/authors/sandrine-zufferey/). Lugano is represented by document-level relation extraction and by abstractive summarisation of product reviews using discourse structure.

### Collaborations

Several people link Swiss institutions to each other. James Henderson appears with both Geneva and Idiap. Lonneke van der Plas is linked to Geneva, Idiap and Lugano. Tanja Samardžić is associated with both Zurich and Geneva, and Sandrine Zufferey and Joel Niklaus with both Bern and Fribourg. The Lake Geneva cluster also shares works. The document-level translation paper and the hypernetwork fine-tuning work both belong to Lausanne and Idiap.

The clearest link between industry and academia is in Zurich. Florian Tramer is connected with both Google (Switzerland) and ETH Zurich, and the company's office sits in the same city as the two Zurich universities. Humanity's Last Exam joins ETH Zurich, Lausanne, Geneva and Bern in a single benchmark effort. Beyond the country, the co-author lists of the shared-task findings for WMT19 and WMT24 tie Swiss researchers to the wider machine translation evaluation community.

### Timeline

From 1990 to 2006 the Swiss works here are few and come mainly from Geneva. They cover test suites for translation evaluation in 1990, translation discovery from non-parallel text in 1995, part-of-speech induction in 2003 and prepositional phrase attachment in 2006.

From 2010 to 2015 the discourse and dialogue strand forms. The ISO dialogue-act work appeared in 2010 and 2012, and connective studies at Idiap and Geneva followed in 2012 and 2013. Cross-lingual parsing came from Zurich in 2011, review summarisation in Lugano in 2014, and the Fribourg discourse papers in 2015.

From 2016 to 2019 neural methods take over. This period brings graph convolutions and biography generation in 2016 and character-level translation in 2017. In 2018 come sentence and word embeddings, document-level translation and the human-parity debate. The 2019 works include the attention-head analysis and context-aware deixis and ellipsis.

From 2020 onward the focus moves to large models and wide-ranging benchmarks. In 2020 there are NumPy, information-theoretic probing, massively multilingual translation and entity linking. The years 2021 to 2023 bring SoundStream, AudioLM, legal benchmarks, reasoning distillation and language–image pretraining. The 2024 work on refusal leads into the 2025 contributions of Humanity's Last Exam and Co-Scientist.

## Where

## Agglomerations (7)

1. [**Zurich**3 institutions*40,737 citations*](https://md.research.elamur.ai/free/en/agglomerations/zurich/)
2. [**Lausanne**1 institution*6,324 citations*](https://md.research.elamur.ai/free/en/agglomerations/lausanne/)
3. [**Geneva**1 institution*2,453 citations*](https://md.research.elamur.ai/free/en/agglomerations/geneva/)
4. [**Lugano**1 institution*639 citations*](https://md.research.elamur.ai/free/en/agglomerations/lugano/)
5. [**Bern**1 institution*486 citations*](https://md.research.elamur.ai/free/en/agglomerations/bern/)
6. [**Sankt Gallen**1 institution*43 citations*](https://md.research.elamur.ai/free/en/agglomerations/st-gallen/)
7. [**Freiburg im Üechtland**1 institution*47 citations*](https://md.research.elamur.ai/free/en/agglomerations/freiburg-im-uechtland/)

## Institutions (10)

### 🎓 Universities (8)

1. **[University of Zurich](https://md.research.elamur.ai/free/en/universities/university-of-zurich/)**University · [Zurich](https://md.research.elamur.ai/free/en/agglomerations/zurich/), Switzerland*44 works · 10,252 citations*
2. **[École Polytechnique Fédérale de Lausanne](https://md.research.elamur.ai/free/en/universities/ecole-polytechnique-federale-de-lausanne/)**University · [Lausanne](https://md.research.elamur.ai/free/en/agglomerations/lausanne/), Switzerland*19 works · 6,324 citations*
3. **[University of Geneva](https://md.research.elamur.ai/free/en/universities/university-of-geneva/)**University · [Geneva](https://md.research.elamur.ai/free/en/agglomerations/geneva/), Switzerland*20 works · 2,453 citations*
4. **[ETH Zurich](https://md.research.elamur.ai/free/en/universities/eth-zurich/)**University · [Zurich](https://md.research.elamur.ai/free/en/agglomerations/zurich/), Switzerland*24 works · 3,273 citations*
5. **[Università della Svizzera italiana](https://md.research.elamur.ai/free/en/universities/universita-della-svizzera-italiana/)**University · [Lugano](https://md.research.elamur.ai/free/en/agglomerations/lugano/), Switzerland*2 works · 639 citations*
6. **[University of Bern](https://md.research.elamur.ai/free/en/universities/university-of-bern/)**University · [Bern](https://md.research.elamur.ai/free/en/agglomerations/bern/), Switzerland*3 works · 486 citations*
7. **[University of St.Gallen](https://md.research.elamur.ai/free/en/universities/university-of-st-gallen/)**University · [Saint Gallen](https://md.research.elamur.ai/free/en/agglomerations/st-gallen/), Switzerland*2 works · 43 citations*
8. **[University of Fribourg](https://md.research.elamur.ai/free/en/universities/university-of-fribourg/)**University · [Fribourg](https://md.research.elamur.ai/free/en/agglomerations/freiburg-im-uechtland/), Switzerland*2 works · 47 citations*

### 🔬 Labs and research institutes (1)

1. **[Idiap Research Institute](https://md.research.elamur.ai/free/en/labs/idiap-research-institute/)**Research institute · Martigny-Combe, Switzerland*17 works · 1,669 citations*

### 🏢 Companies (1)

1. **[Google (Switzerland)](https://md.research.elamur.ai/free/en/companies/google-switzerland/)**Company · [Zurich](https://md.research.elamur.ai/free/en/agglomerations/zurich/), Switzerland*8 works · 27,212 citations*

## People (71)

Authors with a page confirmed at its institutions, the most cited first.

1. [**Caglar Gulcehre**École Polytechnique Fédérale de Lausanne*4 papers here · 30,888 citations*](https://md.research.elamur.ai/free/en/authors/caglar-gulcehre/)
2. [**Rico Sennrich**University of Zurich*23 papers here · 8,189 citations*](https://md.research.elamur.ai/free/en/authors/rico-sennrich/)
3. [**Basil Mustafa**Google (Switzerland)*2 papers here · 7,617 citations*](https://md.research.elamur.ai/free/en/authors/basil-mustafa/)
4. [**Julian Eisenschlos**Google (Switzerland)*2 papers here · 7,097 citations*](https://md.research.elamur.ai/free/en/authors/julian-eisenschlos/)
5. [**Juraj Gottweis**Google (Switzerland)*2 papers here · 6,867 citations*](https://md.research.elamur.ai/free/en/authors/juraj-gottweis/)
6. [**Antoine Bosselut**École Polytechnique Fédérale de Lausanne*6 papers here · 6,472 citations*](https://md.research.elamur.ai/free/en/authors/antoine-bosselut/)
7. [**Florian Tramer**Google (Switzerland)*3 papers here · 5,354 citations*](https://md.research.elamur.ai/free/en/authors/florian-tramer/)
8. [**Marco Tagliasacchi**Google (Switzerland)*2 papers here · 4,037 citations*](https://md.research.elamur.ai/free/en/authors/marco-tagliasacchi/)
9. [**Ryan Cotterell**ETH Zurich*12 papers here · 3,958 citations*](https://md.research.elamur.ai/free/en/authors/ryan-cotterell/)
10. [**Keith Hall**Google (Switzerland)*7 papers here · 3,575 citations*](https://md.research.elamur.ai/free/en/authors/keith-hall/)
11. [**Damien Vincent**Google (Switzerland)*2 papers here · 3,310 citations*](https://md.research.elamur.ai/free/en/authors/damien-vincent/)
12. [**Thomas Hofmann**ETH Zurich*2 papers here · 2,728 citations*](https://md.research.elamur.ai/free/en/authors/thomas-hofmann/)
13. [**Mark Fishel**University of Zurich*8 papers here · 2,498 citations*](https://md.research.elamur.ai/free/en/authors/mark-fishel/)
14. [**Prakhar Gupta**École Polytechnique Fédérale de Lausanne*2 papers here · 2,483 citations*](https://md.research.elamur.ai/free/en/authors/prakhar-gupta/)
15. [**Kumar Shridhar**ETH Zurich*2 papers here · 2,379 citations*](https://md.research.elamur.ai/free/en/authors/kumar-shridhar/)
16. [**Xavier Bresson**École Polytechnique Fédérale de Lausanne*2 papers here · 2,103 citations*](https://md.research.elamur.ai/free/en/authors/xavier-bresson/)
17. [**Lucas Beyer**Google (Switzerland)*3 papers here · 1,986 citations*](https://md.research.elamur.ai/free/en/authors/lucas-beyer/)
18. [**Xiaohua Zhai**Google (Switzerland)*2 papers here · 1,986 citations*](https://md.research.elamur.ai/free/en/authors/xiaohua-zhai/)
19. [**Mathias Müller**University of Zurich*4 papers here · 1,974 citations*](https://md.research.elamur.ai/free/en/authors/mathias-muller/)
20. [**Martin Jaggi**École Polytechnique Fédérale de Lausanne*3 papers here · 1,675 citations*](https://md.research.elamur.ai/free/en/authors/martin-jaggi/)
21. [**Mrinmaya Sachan**ETH Zurich*7 papers here · 1,669 citations*](https://md.research.elamur.ai/free/en/authors/mrinmaya-sachan/)
22. [**Massimiliano Ciaramita**Google (Switzerland)*5 papers here · 1,658 citations*](https://md.research.elamur.ai/free/en/authors/massimiliano-ciaramita/)
23. [**Annette Rios**University of Zurich*4 papers here · 1,435 citations*](https://md.research.elamur.ai/free/en/authors/annette-rios/)
24. [**Matteo Pagliardini**École Polytechnique Fédérale de Lausanne*2 papers here · 1,214 citations*](https://md.research.elamur.ai/free/en/authors/matteo-pagliardini/)
25. [**Yves Scherrer**University of Geneva*4 papers here · 1,177 citations*](https://md.research.elamur.ai/free/en/authors/yves-scherrer/)

[See all 71 people →](https://md.research.elamur.ai/free/en/countries/switzerland/people/)

## Works in our library (123)

Works of its institutions, the most cited first.

1. [**Array Programming with NumPy**Charles R. Harris, K. Jarrod Millman, Stéfan J. van der Walt, et al.*2020 · 23,591 citations*](https://md.research.elamur.ai/free/en/papers/harris-2020-array-programming-numpy/)
2. [**Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering**Michael Defferrard, Xavier Bresson, Pierre Vandergheynst*2016 · 1,797 citations*](https://md.research.elamur.ai/free/en/papers/defferrard-2016-convolutional-neural-networks-graphs-fast/)
3. [**ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks**Fabrizio Gilardi, Meysam Alizadeh, Maël Kubli*2023 · 1,532 citations*](https://md.research.elamur.ai/free/en/papers/gilardi-2023-chatgpt-outperforms-crowd-workers-text/)
4. [**Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned**Elena Voita, David Talbot, Fedor Moiseev, et al.*2019 · 1,511 citations*](https://md.research.elamur.ai/free/en/papers/voita-2019-analyzing-multi-head-self-attention/)
5. [**Sigmoid Loss for Language Image Pre-Training**Xiaohua Zhai, Basil Mustafa, Alexander Kolesnikov, et al.*2023 · 1,427 citations*](https://md.research.elamur.ai/free/en/papers/zhai-2023-sigmoid-loss-language-image-pre/)
6. [**Adapted Large Language Models Can Outperform Medical Experts in Clinical Text Summarization**Dave Van Veen, Cara Van Uden, Louis Blankemeier, et al.*2023 · 925 citations*](https://md.research.elamur.ai/free/en/papers/veen-2023-adapted-large-language-models-outperform/)
7. [**Unsupervised Learning of Sentence Embeddings Using Compositional n-Gram Features**Matteo Pagliardini, Prakhar Gupta, Martin Jaggi*2018 · 840 citations*](https://md.research.elamur.ai/free/en/papers/pagliardini-2018-unsupervised-learning-sentence-embeddings-compositional/)
8. [**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/)
9. [**Colorless Green Recurrent Networks Dream Hierarchically**Kristina Gulordava, Piotr Bojanowski, Edouard Grave, et al.*2018 · 759 citations*](https://md.research.elamur.ai/free/en/papers/gulordava-2018-colorless-green-recurrent-networks-dream/)
10. [**SoundStream: An End-to-End Neural Audio Codec**Neil Zeghidour, Alejandro Luebs, Ahmed Omran, et al.*2021 · 727 citations*](https://md.research.elamur.ai/free/en/papers/zeghidour-2021-soundstream-end-end-neural-audio/)
11. [**Neural Text Generation from Structured Data with Application to the Biography Domain**Remi Lebret, David Grangier, Michael Auli*2016 · 696 citations*](https://md.research.elamur.ai/free/en/papers/lebret-2016-neural-text-generation-structured-data/)
12. [**Learning Word Vectors for 157 Languages**Edouard Grave, Piotr Bojanowski, Prakhar Gupta, et al.*2018 · 640 citations*](https://md.research.elamur.ai/free/en/papers/grave-2018-learning-word-vectors-157-languages/)
13. [**Scalable Zero-shot Entity Linking with Dense Entity Retrieval**Ledell Wu, Fabio Petroni, Martin Josifoski, et al.*2020 · 618 citations*](https://md.research.elamur.ai/free/en/papers/wu-2020-scalable-zero-shot-entity-linking/)
14. [**AudioLM: a Language Modeling Approach to Audio Generation**Zalan Borsos, Raphael Marinier, Damien Vincent, et al.*2022 · 570 citations*](https://md.research.elamur.ai/free/en/papers/borsos-2022-audiolm-language-modeling-approach-audio/)
15. [**Fully Character-Level Neural Machine Translation without Explicit Segmentation**Jason Lee, Kyunghyun Cho, Thomas Hofmann*2017 · 545 citations*](https://md.research.elamur.ai/free/en/papers/lee-2017-fully-character-level-neural-machine/)
16. [**Multi-Source Transfer of Delexicalized Dependency Parsers**Ryan McDonald, Slav Petrov, Keith Hall*2011 · 490 citations*](https://md.research.elamur.ai/free/en/papers/mcdonald-2011-multi-source-transfer-delexicalized-dependency/)
17. [**Improving Massively Multilingual Neural Machine Translation and Zero-Shot Translation**Biao Zhang, Philip Williams, Ivan Titov, et al.*2020 · 472 citations*](https://md.research.elamur.ai/free/en/papers/zhang-2020-improving-massively-multilingual-neural-machine/)
18. [**Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets**Julia Kreutzer, Isaac Caswell, Lisa Wang, et al.*2021 · 447 citations*](https://md.research.elamur.ai/free/en/papers/kreutzer-2021-quality-glance-audit-web-crawled/)
19. [**Context-Aware Neural Machine Translation Learns Anaphora Resolution**Elena Voita, Pavel Serdyukov, Rico Sennrich, et al.*2018 · 446 citations*](https://md.research.elamur.ai/free/en/papers/voita-2018-context-aware-neural-machine-translation/)
20. [**Has Machine Translation Achieved Human Parity? A Case for Document-level Evaluation**Samuel Läubli, Rico Sennrich, Martin Volk*2018 · 443 citations*](https://md.research.elamur.ai/free/en/papers/laubli-2018-has-machine-translation-achieved-human/)
21. [**Document-Level Neural Machine Translation with Hierarchical Attention Networks**Lesly Miculicich, Dhananjay Ram, Nikolaos Pappas, et al.*2018 · 411 citations*](https://md.research.elamur.ai/free/en/papers/miculicich-2018-document-level-neural-machine-translation/)
22. [**Identifying Word Translations in Non-Parallel Texts**Reinhard Rapp*1995 · 394 citations*](https://md.research.elamur.ai/free/en/papers/rapp-1995-identifying-word-translations-non-parallel/)
23. [**Reasoning with Latent Structure Refinement for Document-Level Relation Extraction**Guoshun Nan, Zhijiang Guo, Ivan Sekulic, et al.*2020 · 379 citations*](https://md.research.elamur.ai/free/en/papers/nan-2020-reasoning-latent-structure-refinement-document/)
24. [**When a Good Translation is Wrong in Context: Context-Aware Machine Translation Improves on Deixis, Ellipsis, and Lexical Cohesion**Elena Voita, Rico Sennrich, Ivan Titov*2019 · 321 citations*](https://md.research.elamur.ai/free/en/papers/voita-2019-when-good-translation-is-wrong/)
25. [**Combining Distributional and Morphological Information for Part of Speech Induction**Alexander Clark*2003 · 295 citations*](https://md.research.elamur.ai/free/en/papers/clark-2003-combining-distributional-morphological-information-part/)

[See all 123 works →](https://md.research.elamur.ai/free/en/countries/switzerland/works/)

## Concepts, shelves and topics

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

💡 Concepts 11[Transformer and attention · 1 work](https://md.research.elamur.ai/free/en/concepts/transformer/) · [Semantic similarity · 1 work](https://md.research.elamur.ai/free/en/concepts/semantic-similarity/) · [Word embeddings · 1 work](https://md.research.elamur.ai/free/en/concepts/word-embeddings/) · [Named entity recognition · 1 work](https://md.research.elamur.ai/free/en/concepts/named-entity-recognition/) · [Dependency parsing · 1 work](https://md.research.elamur.ai/free/en/concepts/dependency-parsing/) · [Coreference resolution · 1 work](https://md.research.elamur.ai/free/en/concepts/coreference-resolution/) · [Part-of-speech tagging · 1 work](https://md.research.elamur.ai/free/en/concepts/part-of-speech-tagging/) · [Minimum description length · 1 work](https://md.research.elamur.ai/free/en/concepts/minimum-description-length/) · [Discourse analysis · 1 work](https://md.research.elamur.ai/free/en/concepts/discourse-analysis/) · [Word sense disambiguation · 1 work](https://md.research.elamur.ai/free/en/concepts/word-sense-disambiguation/) · [Psycholinguistics · 2 works](https://md.research.elamur.ai/free/en/concepts/psycholinguistics/)

📚 Shelves 14[Natural language processing · 43 works](https://md.research.elamur.ai/free/en/shelves/nlp/) · [Linguistics & languages · 67 works](https://md.research.elamur.ai/free/en/shelves/linguistics/) · [Other · 11 works](https://md.research.elamur.ai/free/en/shelves/other/) · [Mathematics & statistics · 3 works](https://md.research.elamur.ai/free/en/shelves/mathematics/) · [Machine translation · 28 works](https://md.research.elamur.ai/free/en/shelves/machine-translation/) · [AI & machine learning · 4 works](https://md.research.elamur.ai/free/en/shelves/ai/) · [English · 14 works](https://md.research.elamur.ai/free/en/shelves/en/) · [Computing & GPUs · 2 works](https://md.research.elamur.ai/free/en/shelves/computing/) · [Law · 3 works](https://md.research.elamur.ai/free/en/shelves/law/) · [Education · 1 work](https://md.research.elamur.ai/free/en/shelves/education/) · [Machine translation · 5 works](https://md.research.elamur.ai/free/en/shelves/mtrans/) · [Modern Hebrew · 2 works](https://md.research.elamur.ai/free/en/shelves/he/) · [History & humanities · 2 works](https://md.research.elamur.ai/free/en/shelves/history/) · [Ancient Hebrew · 2 works](https://md.research.elamur.ai/free/en/shelves/hbo/)

🧭 Topics 32[Machine translation · 39 works](https://md.research.elamur.ai/free/en/topics/machine-translation/) · [Evaluation & benchmarks · 22 works](https://md.research.elamur.ai/free/en/topics/evaluation-benchmarks/) · [Language models & pretraining · 7 works](https://md.research.elamur.ai/free/en/topics/language-models/) · [Multilingual & cross-lingual NLP · 8 works](https://md.research.elamur.ai/free/en/topics/multilingual-nlp/) · [Discourse analysis · 10 works](https://md.research.elamur.ai/free/en/topics/discourse-analysis/) · [Speech & spoken language · 5 works](https://md.research.elamur.ai/free/en/topics/speech/) · [Reasoning · 6 works](https://md.research.elamur.ai/free/en/topics/reasoning/) · [Summarization · 2 works](https://md.research.elamur.ai/free/en/topics/summarization/) · [Dependency parsing · 4 works](https://md.research.elamur.ai/free/en/topics/dependency-parsing/) · [Morphology · 4 works](https://md.research.elamur.ai/free/en/topics/morphology/) · [Coreference resolution · 7 works](https://md.research.elamur.ai/free/en/topics/coreference-resolution/) · [Dialogue systems · 4 works](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) · [Text generation · 1 work](https://md.research.elamur.ai/free/en/topics/text-generation/) · [Part-of-speech tagging · 2 works](https://md.research.elamur.ai/free/en/topics/part-of-speech-tagging/) · [Speech act theory · 3 works](https://md.research.elamur.ai/free/en/topics/speech-act-theory/)+17 more

## Works by year

### 2020s

[**2026** 9 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2026/) · [**2025** 8 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2025/) · [**2024** 8 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2024/) · [**2023** 15 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2023/) · [**2022** 8 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2022/) · [**2021** 9 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2021/) · [**2020** 14 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2020/)

### 2010s

[**2019** 6 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2019/) · [**2018** 10 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2018/) · [**2017** 10 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2017/) · [**2016** 2 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2016/) · [**2015** 2 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2015/) · [**2014** 2 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2014/) · [**2013** 5 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2013/) · [**2012** 5 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2012/) · [**2011** 2 works](https://md.research.elamur.ai/free/en/countries/switzerland/year/2011/) · [**2010** 1 work](https://md.research.elamur.ai/free/en/papers/bunt-2010-iso-standard-dialogue-act-annotation/)

### 2000s

[**2009** 1 work](https://md.research.elamur.ai/free/en/papers/merlo-2009-abstraction-generalisation-semantic-role-labels/) · [**2006** 1 work](https://md.research.elamur.ai/free/en/papers/merlo-2006-notion-argument-prepositional-phrase-attachment/) · [**2004** 1 work](https://md.research.elamur.ai/free/en/papers/clark-2004-multi-level-dialogue-act-tags/) · [**2003** 1 work](https://md.research.elamur.ai/free/en/papers/clark-2003-combining-distributional-morphological-information-part/) · [**2002** 1 work](https://md.research.elamur.ai/free/en/papers/volk-2002-combining-unsupervised-supervised-methods-pp/)

### 1990s

[**1995** 1 work](https://md.research.elamur.ai/free/en/papers/rapp-1995-identifying-word-translations-non-parallel/) · [**1990** 1 work](https://md.research.elamur.ai/free/en/papers/king-1990-test-suites-evaluation-machine-translation/)

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

Updated 23 October 2026