% $ biblatex auxiliary file $
% $ biblatex bbl format version 3.3 $
% Do not modify the above lines!
%
% This is an auxiliary file used by the 'biblatex' package.
% This file may safely be deleted. It will be recreated by
% biber as required.
%
\begingroup
\makeatletter
\@ifundefined{ver@biblatex.sty}
  {\@latex@error
     {Missing 'biblatex' package}
     {The bibliography requires the 'biblatex' package.}
      \aftergroup\endinput}
  {}
\endgroup


\refsection{1}
  \datalist[entry]{nty/global//global/global/global}
    \entry{kokosinskii2024substitution}{inproceedings}{}{}
      \name{author}{2}{}{%
        {{hash=037065b6981381a88a92af47594a24dc}{%
           family={Kokosinskii},
           familyi={K\bibinitperiod},
           given={Denis},
           giveni={D\bibinitperiod}}}%
        {{hash=75239b75cdb093d1adfdec438377fbac}{%
           family={Arefyev},
           familyi={A\bibinitperiod},
           given={Nikolay},
           giveni={N\bibinitperiod}}}%
      }
      \name{editor}{6}{}{%
        {{hash=c341e33a4b6ea903f6c723a46650c1be}{%
           family={Calzolari},
           familyi={C\bibinitperiod},
           given={Nicoletta},
           giveni={N\bibinitperiod}}}%
        {{hash=549f49a0f40722c52169e0083730d12f}{%
           family={Kan},
           familyi={K\bibinitperiod},
           given={Min-Yen},
           giveni={M\bibinithyphendelim Y\bibinitperiod}}}%
        {{hash=a2552991b1433efe99cce9b2a665aae7}{%
           family={Hoste},
           familyi={H\bibinitperiod},
           given={Veronique},
           giveni={V\bibinitperiod}}}%
        {{hash=5ca65e378c2b71d06107979d273315ed}{%
           family={Lenci},
           familyi={L\bibinitperiod},
           given={Alessandro},
           giveni={A\bibinitperiod}}}%
        {{hash=8f51dd1b39e7ef8aa830c97a98536583}{%
           family={Sakti},
           familyi={S\bibinitperiod},
           given={Sakriani},
           giveni={S\bibinitperiod}}}%
        {{hash=47c3156f10dfabf6842e8caf450fda15}{%
           family={Xue},
           familyi={X\bibinitperiod},
           given={Nianwen},
           giveni={N\bibinitperiod}}}%
      }
      \list{location}{1}{%
        {Torino, Italia}%
      }
      \list{publisher}{1}{%
        {ELRA and ICCL}%
      }
      \strng{namehash}{cd421eea76fd319b2d3a189b0149d8e3}
      \strng{fullhash}{b7168eecd43e294f5747276b1d7816ce}
      \strng{fullhashraw}{b7168eecd43e294f5747276b1d7816ce}
      \strng{bibnamehash}{b7168eecd43e294f5747276b1d7816ce}
      \strng{authorbibnamehash}{b7168eecd43e294f5747276b1d7816ce}
      \strng{authornamehash}{cd421eea76fd319b2d3a189b0149d8e3}
      \strng{authorfullhash}{b7168eecd43e294f5747276b1d7816ce}
      \strng{authorfullhashraw}{b7168eecd43e294f5747276b1d7816ce}
      \strng{editorbibnamehash}{c0f71bf98493a021e0ad741a5193b2ff}
      \strng{editornamehash}{c0f71bf98493a021e0ad741a5193b2ff}
      \strng{editorfullhash}{67cc341797d15e0b10b5e05c259573e8}
      \strng{editorfullhashraw}{67cc341797d15e0b10b5e05c259573e8}
      \field{sortinit}{K}
      \field{sortinithash}{c02bf6bff1c488450c352b40f5d853ab}
      \field{labelnamesource}{author}
      \field{labeltitlesource}{title}
      \field{abstract}{Word Sense Induction (WSI) is the task of discovering senses of an ambiguous word by grouping usages of this word into clusters corresponding to these senses. Many approaches were proposed to solve WSI in English and a few other languages, but these approaches are not easily adaptable to new languages. We present multilingual substitution-based WSI methods that support any of 100 languages covered by the underlying multilingual language model with minimal to no adaptation required. Despite the multilingual capabilities, our methods perform on par with the existing monolingual approaches on popular English WSI datasets. At the same time, they will be most useful for lower-resourced languages which miss lexical resources available for English, thus, have higher demand for unsupervised methods like WSI.}
      \field{booktitle}{Proceedings of the 2024 {{Joint International Conference}} on {{Computational Linguistics}}, {{Language Resources}} and {{Evaluation}} ({{LREC-COLING}} 2024)}
      \field{month}{5}
      \field{title}{Multilingual {{Substitution-based Word Sense Induction}}}
      \field{urlday}{26}
      \field{urlmonth}{5}
      \field{urlyear}{2026}
      \field{year}{2024}
      \field{urldateera}{ce}
      \field{pages}{11859\bibrangedash 11872}
      \range{pages}{14}
      \verb{file}
      \verb /Users/pablo/Zotero/storage/8Z927Z57/Kokosinskii and Arefyev - Multilingual Substitution-based Word Sense Induction.pdf;/Users/pablo/Zotero/storage/JDJ2IUSQ/Kokosinskii and Arefyev - 2024 - Multilingual Substitution-based Word Sense Induction.pdf
      \endverb
    \endentry
    \entry{xlwsd}{article}{}{}
      \name{author}{3}{}{%
        {{hash=9735617519e4929991b4c08b1d7f37ea}{%
           family={Pasini},
           familyi={P\bibinitperiod},
           given={Tommaso},
           giveni={T\bibinitperiod}}}%
        {{hash=020f393f61fcfa517b4961ac7b598459}{%
           family={Raganato},
           familyi={R\bibinitperiod},
           given={Alessandro},
           giveni={A\bibinitperiod}}}%
        {{hash=6fe24dce481a67e247581d4b20435709}{%
           family={Navigli},
           familyi={N\bibinitperiod},
           given={Roberto},
           giveni={R\bibinitperiod}}}%
      }
      \strng{namehash}{4a54531b34123e4c7d1ea88f8e5f99cc}
      \strng{fullhash}{a43111b992d0c3f97fd8bc05c64ce054}
      \strng{fullhashraw}{a43111b992d0c3f97fd8bc05c64ce054}
      \strng{bibnamehash}{a43111b992d0c3f97fd8bc05c64ce054}
      \strng{authorbibnamehash}{a43111b992d0c3f97fd8bc05c64ce054}
      \strng{authornamehash}{4a54531b34123e4c7d1ea88f8e5f99cc}
      \strng{authorfullhash}{a43111b992d0c3f97fd8bc05c64ce054}
      \strng{authorfullhashraw}{a43111b992d0c3f97fd8bc05c64ce054}
      \field{sortinit}{P}
      \field{sortinithash}{ff3bcf24f47321b42cb156c2cc8a8422}
      \field{labelnamesource}{author}
      \field{labeltitlesource}{shorttitle}
      \field{abstract}{Transformer-based architectures brought a breeze of change to Word Sense Disambiguation (WSD), improving models' performances by a large margin. The fast development of new approaches has been further encouraged by a well-framed evaluation suite for English, which has allowed their performances to be kept track of and compared fairly. However, other languages have remained largely unexplored, as testing data are available for a few languages only and the evaluation setting is rather matted. In this paper, we untangle this situation by proposing XL-WSD, a cross-lingual evaluation benchmark for the WSD task featuring sense-annotated development and test sets in 18 languages from six different linguistic families, together with language-specific silver training data. We leverage XL-WSD datasets to conduct an extensive evaluation of neural and knowledge-based approaches, including the most recent multilingual language models. Results show that the zero-shot knowledge transfer across languages is a promising research direction within the WSD field, especially when considering low-resourced languages where large pretrained multilingual models still perform poorly. We make the evaluation suite and the code for performing the experiments available at https://sapienzanlp.github.io/xl-wsd/.}
      \field{issn}{2374-3468, 2159-5399}
      \field{journaltitle}{Proceedings of the AAAI Conference on Artificial Intelligence}
      \field{langid}{english}
      \field{month}{5}
      \field{number}{15}
      \field{shorttitle}{{{XL-WSD}}}
      \field{title}{{{XL-WSD}}: {{An Extra-Large}} and {{Cross-Lingual Evaluation Framework}} for {{Word Sense Disambiguation}}}
      \field{urlday}{7}
      \field{urlmonth}{6}
      \field{urlyear}{2026}
      \field{volume}{35}
      \field{year}{2021}
      \field{urldateera}{ce}
      \field{pages}{13648\bibrangedash 13656}
      \range{pages}{9}
      \verb{doi}
      \verb 10.1609/aaai.v35i15.17609
      \endverb
      \verb{file}
      \verb /Users/pablo/Zotero/storage/35UVFLZV/Pasini et al. - 2021 - XL-WSD An Extra-Large and Cross-Lingual Evaluation Framework for Word Sense Disambiguation.pdf
      \endverb
    \endentry
    \entry{rother-2020}{inproceedings}{}{}
      \name{author}{3}{}{%
        {{hash=ac23da37927d0ed86dc8c0993b02e9a1}{%
           family={Rother},
           familyi={R\bibinitperiod},
           given={David},
           giveni={D\bibinitperiod}}}%
        {{hash=0f58441768e133dd0f2872c7f264c16f}{%
           family={Haider},
           familyi={H\bibinitperiod},
           given={Thomas},
           giveni={T\bibinitperiod}}}%
        {{hash=fa4a8f2209aafcafe807835e7d616b7b}{%
           family={Eger},
           familyi={E\bibinitperiod},
           given={Steffen},
           giveni={S\bibinitperiod}}}%
      }
      \name{editor}{6}{}{%
        {{hash=b904bf2b776e31eeda30178096489a72}{%
           family={Herbelot},
           familyi={H\bibinitperiod},
           given={Aurelie},
           giveni={A\bibinitperiod}}}%
        {{hash=7db206795c9c6671b7060b6153bcdab6}{%
           family={Zhu},
           familyi={Z\bibinitperiod},
           given={Xiaodan},
           giveni={X\bibinitperiod}}}%
        {{hash=f4824cba97513223c504aa4a606948ad}{%
           family={Palmer},
           familyi={P\bibinitperiod},
           given={Alexis},
           giveni={A\bibinitperiod}}}%
        {{hash=989a024048b92d0c851aa77e90dfa766}{%
           family={Schneider},
           familyi={S\bibinitperiod},
           given={Nathan},
           giveni={N\bibinitperiod}}}%
        {{hash=be5886b6b904bf38c0617ff1305a5300}{%
           family={May},
           familyi={M\bibinitperiod},
           given={Jonathan},
           giveni={J\bibinitperiod}}}%
        {{hash=c36f9fc7c3f7617b9ee71cb8646c6f30}{%
           family={Shutova},
           familyi={S\bibinitperiod},
           given={Ekaterina},
           giveni={E\bibinitperiod}}}%
      }
      \list{location}{1}{%
        {Barcelona (online)}%
      }
      \list{publisher}{1}{%
        {International Committee for Computational Linguistics}%
      }
      \strng{namehash}{f39c265ad978bf9fcaf4f9511011ed2f}
      \strng{fullhash}{21a7f9d86b565fca9445abc60b364a88}
      \strng{fullhashraw}{21a7f9d86b565fca9445abc60b364a88}
      \strng{bibnamehash}{21a7f9d86b565fca9445abc60b364a88}
      \strng{authorbibnamehash}{21a7f9d86b565fca9445abc60b364a88}
      \strng{authornamehash}{f39c265ad978bf9fcaf4f9511011ed2f}
      \strng{authorfullhash}{21a7f9d86b565fca9445abc60b364a88}
      \strng{authorfullhashraw}{21a7f9d86b565fca9445abc60b364a88}
      \strng{editorbibnamehash}{2428a80177eb2ea6ec63aceb19a75446}
      \strng{editornamehash}{2428a80177eb2ea6ec63aceb19a75446}
      \strng{editorfullhash}{1ecf97f89986e570e357f53d78c0ea4c}
      \strng{editorfullhashraw}{1ecf97f89986e570e357f53d78c0ea4c}
      \field{sortinit}{R}
      \field{sortinithash}{5e1c39a9d46ffb6bebd8f801023a9486}
      \field{labelnamesource}{author}
      \field{labeltitlesource}{shorttitle}
      \field{abstract}{This paper describes the system Clustering on Manifolds of Contextualized Embeddings (CMCE) submitted to the SemEval-2020 Task 1 on Unsupervised Lexical Semantic Change Detection. Subtask 1 asks to identify whether or not a word gained/lost a sense across two time periods. Subtask 2 is about computing a ranking of words according to the amount of change their senses underwent. Our system uses contextualized word embeddings from MBERT, whose dimensionality we reduce with an autoencoder and the UMAP algorithm, to be able to use a wider array of clustering algorithms that can automatically determine the number of clusters. We use Hierarchical Density Based Clustering (HDBSCAN) and compare it to Gaussian MixtureModels (GMMs) and other clustering algorithms. Remarkably, with only 10 dimensional MBERT embeddings (reduced from the original size of 768), our submitted model performs best on subtask 1 for English and ranks third in subtask 2 for English. In addition to describing our system, we discuss our hyperparameter configurations and examine why our system lags behind for the other languages involved in the shared task (German, Swedish, Latin). Our code is available at https://github.com/DavidRother/semeval2020-task1}
      \field{booktitle}{Proceedings of the {{Fourteenth Workshop}} on {{Semantic Evaluation}}}
      \field{month}{12}
      \field{shorttitle}{{{CMCE}} at {{SemEval-2020 Task}} 1}
      \field{title}{{{CMCE}} at {{SemEval-2020 Task}} 1: {{Clustering}} on {{Manifolds}} of {{Contextualized Embeddings}} to {{Detect Historical Meaning Shifts}}}
      \field{urlday}{25}
      \field{urlmonth}{4}
      \field{urlyear}{2026}
      \field{year}{2020}
      \field{urldateera}{ce}
      \field{pages}{187\bibrangedash 193}
      \range{pages}{7}
      \verb{doi}
      \verb 10.18653/v1/2020.semeval-1.22
      \endverb
      \verb{file}
      \verb /Users/pablo/Zotero/storage/R865BGX8/Rother et al. - 2020 - CMCE at SemEval-2020 Task 1 Clustering on Manifolds of Contextualized Embeddings to Detect Historic.pdf
      \endverb
    \endentry
    \entry{wiedemann2019}{article}{}{}
      \name{author}{4}{}{%
        {{hash=99bc941a4e758b1c84535f8fa482e2ec}{%
           family={Wiedemann},
           familyi={W\bibinitperiod},
           given={Gregor},
           giveni={G\bibinitperiod}}}%
        {{hash=7d4436b9e47ef1a7808c3135fd774c1b}{%
           family={Remus},
           familyi={R\bibinitperiod},
           given={Steffen},
           giveni={S\bibinitperiod}}}%
        {{hash=046f74a61eef4518f3e3b74a99297178}{%
           family={Chawla},
           familyi={C\bibinitperiod},
           given={Avi},
           giveni={A\bibinitperiod}}}%
        {{hash=786eca9f4966307b17cae6c3bba98905}{%
           family={Biemann},
           familyi={B\bibinitperiod},
           given={Chris},
           giveni={C\bibinitperiod}}}%
      }
      \strng{namehash}{9690c5ae0618e1db1e956c585af85c5f}
      \strng{fullhash}{e31ab4fbff1552fa8d70c823ff681cc8}
      \strng{fullhashraw}{e31ab4fbff1552fa8d70c823ff681cc8}
      \strng{bibnamehash}{9690c5ae0618e1db1e956c585af85c5f}
      \strng{authorbibnamehash}{9690c5ae0618e1db1e956c585af85c5f}
      \strng{authornamehash}{9690c5ae0618e1db1e956c585af85c5f}
      \strng{authorfullhash}{e31ab4fbff1552fa8d70c823ff681cc8}
      \strng{authorfullhashraw}{e31ab4fbff1552fa8d70c823ff681cc8}
      \field{sortinit}{W}
      \field{sortinithash}{4315d78024d0cea9b57a0c6f0e35ed0d}
      \field{labelnamesource}{author}
      \field{labeltitlesource}{title}
      \field{abstract}{Contextualized word embeddings (CWE) such as provided by ELMo (Peters et al., 2018), Flair NLP (Akbik et al., 2018), or BERT (Devlin et al., 2019) are a major recent innovation in NLP. CWEs provide semantic vector representations of words depending on their respective context. Their advantage over static word embeddings has been shown for a number of tasks, such as text classification, sequence tagging, or machine translation. Since vectors of the same word type can vary depending on the respective context, they implicitly provide a model for word sense disambiguation (WSD). We introduce a simple but effective approach to WSD using a nearest neighbor classification on CWEs. We compare the performance of different CWE models for the task and can report improvements above the current state of the art for two standard WSD benchmark datasets. We further show that the pre-trained BERT model is able to place polysemic words into distinct `sense' regions of the embedding space, while ELMo and Flair NLP do not seem to possess this ability.}
      \field{langid}{english}
      \field{title}{Does {{BERT Make Any Sense}}? {{Interpretable Word Sense Disambiguation}} with {{Contextualized Embeddings}}}
      \field{year}{2019}
      \verb{file}
      \verb /Users/pablo/Zotero/storage/FY8KANZU/Wiedemann et al. - Does BERT Make Any Sense Interpretable Word Sense Disambiguation with Contextualized Embeddings.pdf
      \endverb
    \endentry
  \enddatalist
\endrefsection
\endinput