Aktualitéiten – Seite 38 – Infolux – Fuerschungsportal iwwert d'Lëtzebuergescht

Päiperlek, Pimpampel & co.

SchnëssenVon Nathalie Entringer 24. Juli 2018 Kommentar hinterlassen

An der Schnëssen-App hu 1459 Leit eis geäntwert a sech opgeholl, wéi si déi kleng, meescht faarweg Déiercher mat de Flilleken nennen an dobäi konnte mir feststellen, datt eng grouss Majoritéit Päiperlek seet. Déi aner zwou nennenswäert Variante si Pimpampel a Schmetterleng. All dës Forme si souguer schonn am Luxemburger Wörterbuch vermierkt gewiescht, woubäi de Schmetter-leng do nach ënnert där héichdäitscher Aussprooch Schmetter-ling ze…

Details

Vamos a la playa: Strand vs. Plage

SchnëssenVon Nathalie Entringer 13. Juli 2018 Kommentar hinterlassen

Egal, ob Dir elo op d’Playa gitt, am Schaukelstull um Beach sëtzt, um Strand danzt, op d’belsch Plage fuert oder eppes maacht, wat glat a guer näischt mat Sand a Mier ze dinn huet, wëlle mir Iech elo schonn eng schéi Summervakanz wënschen! Déi schmock Uspillung un Iech sécherlech bekannte Lidder kënnt dobäi net vun ongeféier…

Details

Eng Monster-Party an der Nuecht

SchnëssenVon Nathalie Entringer 6. Juli 2018 Kommentar hinterlassen

Gitt uecht! ‚t ass Mëtternuecht! Gitt uecht! ‚t ass däischter um Späicher an d’Geeschter si Meeschter. Nee, et geet elo hei net, wéi de Ufank vum Jhemp Hoscheit sengem Gedicht „Eng Monster-Party“ gleewe loosse kéint ëm Geeschter, Hexen a Piraten, ma et geet ëm déi Zäit an där sech dës Wiese ganz wuel fillen – et geet ëm d’Nuecht. 858…

Details

De klenge ronne Fall aus Lieder: E Futtball, e Fussball oder e Foussball?

SchnëssenVon Nathalie Entringer 17. Juni 2018 Kommentar hinterlassen

Vill Leit sinn de Moment jo am WM-Féiwer an déi eng oder aner vun Iech hate bestëmmt schonn eng Diskussioun an der Famill, mat Frënn oder Kolleegen doriwwer, ob et dann elo Fussball, Foussball oder Futtball heescht. Mir hunn d’Äntwert oder op d’mannst eng éischt Auswäertung vun deem, wat Dir eis beim Bild vum ronne Lieder an d’App erageschwat hutt.…

Details

Iwwer dat oranget Geméis: Muerten a Wuerzelen

SchnëssenVon Nathalie Entringer 14. Juni 2018 Kommentar hinterlassen

Nodeems mir eis 1087 Opnamen zum Bild mam orangë Geméis ugelauschtert hunn, kënne mir Iech folgend flott éischt Resultater presentéieren:Et ginn 2 Haaptvarianten: Muert a Wuerzel. Muert ass vun insgesamt 62,2 % (n=669) vun de Leit gesot ginn a Wuerzel kënnt zu 28,1 % (n=302) vir. Déi aner 9,7 % verdeelen sech op weider Variante vu Muert (Morten, Murschten a Muurten), Karotten (1,4%), Duebelnennungen a vereenzelt aner…

Details

Neist Publikatiounen am BiblioLux

56201 1 chicago-fullnote-bibliography 5 date desc 7910 https://infolux.uni.lu/wp-content/plugins/zotpress/

%7B%22status%22%3A%22success%22%2C%22updateneeded%22%3Afalse%2C%22instance%22%3Afalse%2C%22meta%22%3A%7B%22request_last%22%3A0%2C%22request_next%22%3A0%2C%22used_cache%22%3Atrue%7D%2C%22data%22%3A%5B%7B%22key%22%3A%227KZKRJ39%22%2C%22library%22%3A%7B%22id%22%3A56201%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Gilles%20and%20Hinskens%22%2C%22parsedDate%22%3A%222026-06-01%22%2C%22numChildren%22%3A0%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BGilles%2C%20Peter%2C%20and%20Frans%20Hinskens.%20%26%23x201C%3BReconstructing%20the%20Spread%20of%20Word-Final%20%5C%2Fn%5C%2F%20Deletion%20in%20Central%20Franconian%20Dialects.%20Internal%20and%20Geographical%20Dynamics.%26%23x201D%3B%20%26lt%3Bi%26gt%3BJournal%20of%20Germanic%20Linguistics%26lt%3B%5C%2Fi%26gt%3B%2C%20June%201%2C%202026%2C%201%26%23x2013%3B41.%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1017%5C%2FS1470542726100208%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.1017%5C%2FS1470542726100208%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22journalArticle%22%2C%22title%22%3A%22Reconstructing%20the%20spread%20of%20word-final%20%5C%2Fn%5C%2F%20deletion%20in%20Central%20Franconian%20dialects.%20Internal%20and%20geographical%20dynamics%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Peter%22%2C%22lastName%22%3A%22Gilles%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Frans%22%2C%22lastName%22%3A%22Hinskens%22%7D%5D%2C%22abstractNote%22%3A%22Abstract%5Cn%20%20%20%20%20%20%20%20%20%20%20%20This%20study%20examines%20the%20phenomenon%20of%20word-final%20%5C%2Fn%5C%2F%20deletion%20%28WFnD%29%20in%20Central%20Franconian%20dialects%20spoken%20across%20parts%20of%20the%20Netherlands%2C%20Belgium%2C%20Luxembourg%2C%20Germany%20and%20France.%20The%20deletion%20process%20occurs%20after%20schwa%20and%20full%20vowels%20and%20manifests%20itself%20in%20different%20phonological%20patterns%20across%20dialects.%20The%20paper%20presents%20a%20comparative%20analysis%20based%20on%20a%20selection%20of%20local%20varieties%2C%20investigating%20the%20effects%20of%20the%20left-hand%20and%20right-hand%20phonological%20context%20on%20WFnD.%20The%20theoretical%20framework%20of%20the%20%5Cu2018Life%20Cycle%20of%20Phonological%20Processes%5Cu2019%20serves%20to%20reconstruct%20the%20diachronic%20development%20of%20WFnD%2C%20allowing%20to%20trace%20its%20progression%20from%20systematic%20phonetic%20variation%20to%20a%20categorical%20phonological%20phenomenon%20and%20lexicalized%20forms.%20By%20applying%20this%20framework%2C%20the%20study%20demonstrates%20how%20different%20dialects%20exhibit%20WFnD%20at%20distinct%20stages%20of%20the%20life%20cycle%2C%20revealing%20the%20systematic%20nature%20of%20its%20historical%20unfolding.%20The%20findings%20highlight%20the%20role%20and%20interplay%20of%20prosodic%2C%20phonological%2C%20morphological%20and%20lexical%20factors%20conditioning%20WFnD.%22%2C%22date%22%3A%222026-06-01%22%2C%22section%22%3A%22%22%2C%22partNumber%22%3A%22%22%2C%22partTitle%22%3A%22%22%2C%22DOI%22%3A%2210.1017%5C%2FS1470542726100208%22%2C%22citationKey%22%3A%22%22%2C%22url%22%3A%22https%3A%5C%2F%5C%2Fwww.cambridge.org%5C%2Fcore%5C%2Fproduct%5C%2Fidentifier%5C%2FS1470542726100208%5C%2Ftype%5C%2Fjournal_article%22%2C%22PMID%22%3A%22%22%2C%22PMCID%22%3A%22%22%2C%22ISSN%22%3A%221470-5427%2C%201475-3014%22%2C%22language%22%3A%22en%22%2C%22collections%22%3A%5B%5D%2C%22dateModified%22%3A%222026-06-01T09%3A17%3A14Z%22%7D%7D%2C%7B%22key%22%3A%222P9NPACH%22%2C%22library%22%3A%7B%22id%22%3A56201%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Milano%20et%20al.%22%2C%22parsedDate%22%3A%222026-04-30%22%2C%22numChildren%22%3A1%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BMilano%2C%20Emilia%2C%20Alistair%20Plum%2C%20Yves%20Scherrer%2C%20and%20Christoph%20Purschke.%20%26%23x201C%3BLanguage%20Ideologies%20in%20a%20Multilingual%20Society%3A%20An%20LLM-Based%20Analysis%20of%20Luxembourgish%20News%20Comments%2C%26%23x201D%3B%20April%2030%2C%202026.%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2604.27661%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2604.27661%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22preprint%22%2C%22title%22%3A%22Language%20Ideologies%20in%20a%20Multilingual%20Society%3A%20An%20LLM-based%20Analysis%20of%20Luxembourgish%20News%20Comments%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Emilia%22%2C%22lastName%22%3A%22Milano%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alistair%22%2C%22lastName%22%3A%22Plum%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Yves%22%2C%22lastName%22%3A%22Scherrer%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Christoph%22%2C%22lastName%22%3A%22Purschke%22%7D%5D%2C%22abstractNote%22%3A%22Detecting%20language%20ideologies%20is%20a%20valuable%20yet%20complex%20task%20for%20understanding%20how%20identities%20are%20constructed%20through%20discourse.%20In%20Luxembourg%26%23039%3Bs%20multicultural%20and%20multilingual%20society%2C%20language%20ideologies%20reflect%20more%20than%20simple%20preferences%3A%20they%20carry%20deep%20cultural%20and%20social%20meanings%2C%20shaping%20identities%20and%20social%20belonging.%20Following%20recent%20developments%20in%20applying%20Natural%20Language%20Processing%20tools%20to%20linguistics%20and%20social%20science%2C%20this%20paper%20explores%20the%20potential%20of%20large%20language%20models%20to%20assist%20in%20the%20detection%20of%20language%20ideologies.%20We%20manually%20annotate%20a%20corpus%20of%20user%20comments%20in%20Luxembourgish%20with%20predefined%20ideological%20categories%20and%20then%20evaluate%20the%20performance%20of%20large%20language%20models%20under%20varying%20prompt%20conditions%20to%20assess%20their%20ability%20to%20replicate%20these%20human%20annotations.%20Since%20Luxembourgish%20is%20a%20small%20language%20and%20poorly%20represented%20in%20the%20LLMs%26%23039%3B%20training%20data%2C%20we%20also%20investigate%20whether%20machine-translating%20the%20data%20to%20high-resource%20languages%20increases%20performance%20on%20the%20ideology%20detection%20task.%20Our%20findings%20suggest%20that%2C%20while%20LLMs%20are%20not%20yet%20fully%20optimized%20for%20a%20multi-class%20ideological%20annotation%20task%2C%20they%20are%20practical%20tools%20to%20identify%20language%20ideological%20content.%22%2C%22genre%22%3A%22%22%2C%22repository%22%3A%22%22%2C%22archiveID%22%3A%22arXiv%3A2604.27661%22%2C%22date%22%3A%222026-04-30%22%2C%22DOI%22%3A%2210.48550%5C%2FarXiv.2604.27661%22%2C%22citationKey%22%3A%22%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Farxiv.org%5C%2Fabs%5C%2F2604.27661%22%2C%22language%22%3A%22%22%2C%22collections%22%3A%5B%5D%2C%22dateModified%22%3A%222026-05-06T12%3A42%3A22Z%22%7D%7D%2C%7B%22key%22%3A%2282E5VD8G%22%2C%22library%22%3A%7B%22id%22%3A56201%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Lutgen%20et%20al.%22%2C%22parsedDate%22%3A%222026-03-25%22%2C%22numChildren%22%3A3%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BLutgen%2C%20Anne-Marie%2C%20Alistair%20Plum%2C%20Verena%20Blaschke%2C%20Barbara%20Plank%2C%20and%20Christoph%20Purschke.%20%26%23x201C%3BVariation%20Is%20the%20Norm%3A%20Embracing%20Sociolinguistics%20in%20NLP.%26%23x201D%3B%20arXiv%2C%20March%2025%2C%202026.%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2603.24222%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2603.24222%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22preprint%22%2C%22title%22%3A%22Variation%20is%20the%20Norm%3A%20Embracing%20Sociolinguistics%20in%20NLP%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Anne-Marie%22%2C%22lastName%22%3A%22Lutgen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alistair%22%2C%22lastName%22%3A%22Plum%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Verena%22%2C%22lastName%22%3A%22Blaschke%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Barbara%22%2C%22lastName%22%3A%22Plank%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Christoph%22%2C%22lastName%22%3A%22Purschke%22%7D%5D%2C%22abstractNote%22%3A%22In%20Natural%20Language%20Processing%20%28NLP%29%2C%20variation%20is%20typically%20seen%20as%20noise%20and%20%26quot%3Bnormalised%20away%26quot%3B%20before%20processing%2C%20even%20though%20it%20is%20an%20integral%20part%20of%20language.%20Conversely%2C%20studying%20language%20variation%20in%20social%20contexts%20is%20central%20to%20sociolinguistics.%20We%20present%20a%20framework%20to%20combine%20the%20sociolinguistic%20dimension%20of%20language%20with%20the%20technical%20dimension%20of%20NLP.%20We%20argue%20that%20by%20embracing%20sociolinguistics%2C%20variation%20can%20actively%20be%20included%20in%20a%20research%20setup%2C%20in%20turn%20informing%20the%20NLP%20side.%20To%20illustrate%20this%2C%20we%20provide%20a%20case%20study%20on%20Luxembourgish%2C%20an%20evolving%20language%20featuring%20a%20large%20amount%20of%20orthographic%20variation%2C%20demonstrating%20how%20NLP%20performance%20is%20impacted.%20The%20results%20show%20large%20discrepancies%20in%20the%20performance%20of%20models%20tested%20and%20fine-tuned%20on%20data%20with%20a%20large%20amount%20of%20orthographic%20variation%20in%20comparison%20to%20data%20closer%20to%20the%20%28orthographic%29%20standard.%20Furthermore%2C%20we%20provide%20a%20possible%20solution%20to%20improve%20the%20performance%20by%20including%20variation%20in%20the%20fine-tuning%20process.%20This%20case%20study%20highlights%20the%20importance%20of%20including%20variation%20in%20the%20research%20setup%2C%20as%20models%20are%20currently%20not%20robust%20to%20occurring%20variation.%20Our%20framework%20facilitates%20the%20inclusion%20of%20variation%20in%20the%20thought-process%20while%20also%20being%20grounded%20in%20the%20theoretical%20framework%20of%20sociolinguistics.%22%2C%22genre%22%3A%22%22%2C%22repository%22%3A%22arXiv%22%2C%22archiveID%22%3A%22arXiv%3A2603.24222%22%2C%22date%22%3A%222026-03-25%22%2C%22DOI%22%3A%2210.48550%5C%2FarXiv.2603.24222%22%2C%22citationKey%22%3A%22%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Farxiv.org%5C%2Fabs%5C%2F2603.24222%22%2C%22language%22%3A%22%22%2C%22collections%22%3A%5B%5D%2C%22dateModified%22%3A%222026-04-01T08%3A23%3A10Z%22%7D%7D%2C%7B%22key%22%3A%2225BZH6E2%22%2C%22library%22%3A%7B%22id%22%3A56201%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Rehlinger%22%2C%22parsedDate%22%3A%222026-02-17%22%2C%22numChildren%22%3A3%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BRehlinger%2C%20Nils.%20%26%23x201C%3BLuxMT%20Technical%20Report.%26%23x201D%3B%20arXiv%2C%20February%2017%2C%202026.%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2602.15506%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2602.15506%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22preprint%22%2C%22title%22%3A%22LuxMT%20Technical%20Report%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Nils%22%2C%22lastName%22%3A%22Rehlinger%22%7D%5D%2C%22abstractNote%22%3A%22We%20introduce%20LuxMT%2C%20a%20machine%20translation%20system%20based%20on%20Gemma%203%2027B%20and%20fine-tuned%20for%20translation%20from%20Luxembourgish%20%28LB%29%20into%20French%20%28FR%29%20and%20English%20%28EN%29.%20To%20assess%20translation%20performance%2C%20we%20construct%20a%20novel%20benchmark%20covering%20LB-FR%2C%20LB-EN%2C%20and%20LB-FR%20using%20human-translated%20data%20from%20Luci%2C%20a%20tourist%20magazine%20about%20Luxembourg.%20Training%20data%20stems%20from%20LuxAlign%2C%20a%20parallel%20corpus%20of%20multilingual%20Luxembourgish%20news%20articles%2C%20and%20LB%20parliamentary%20transcripts%20augmented%20with%20Google%20Translate.%20We%20filter%20the%20data%20using%20LuxEmbedder%2C%20LB%20sentence%20embeddings%2C%20to%20remove%20low-equivalence%20segment-pairs.%20Overall%2C%20LuxMT%26%23039%3Bs%20results%20suggest%20strong%20improvements%20over%20the%20Gemma%203%20baseline%2C%20even%20for%20translating%20LB%20to%20German%20%28DE%29%2C%20despite%20the%20training%20data%20not%20containing%20any%20DE.%20We%20also%20explore%20LuxEmbedder%26%23039%3Bs%20potential%20to%20be%20used%20as%20a%20quality%20estimation%20metric%20and%20find%20strong%20correlations%20with%20other%20reference-based%20metrics.%20However%2C%20we%20call%20for%20further%20research%20to%20fully%20assess%20the%20metric%26%23039%3Bs%20utility%20and%20advise%20using%20it%20with%20caution.%22%2C%22genre%22%3A%22%22%2C%22repository%22%3A%22arXiv%22%2C%22archiveID%22%3A%22arXiv%3A2602.15506%22%2C%22date%22%3A%222026-02-17%22%2C%22DOI%22%3A%2210.48550%5C%2FarXiv.2602.15506%22%2C%22citationKey%22%3A%22%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Farxiv.org%5C%2Fabs%5C%2F2602.15506%22%2C%22language%22%3A%22%22%2C%22collections%22%3A%5B%5D%2C%22dateModified%22%3A%222026-02-20T12%3A32%3A54Z%22%7D%7D%2C%7B%22key%22%3A%22F6DGLREI%22%2C%22library%22%3A%7B%22id%22%3A56201%7D%2C%22meta%22%3A%7B%22creatorSummary%22%3A%22Lutgen%20et%20al.%22%2C%22parsedDate%22%3A%222026-02-12%22%2C%22numChildren%22%3A2%7D%2C%22bib%22%3A%22%26lt%3Bdiv%20class%3D%26quot%3Bcsl-bib-body%26quot%3B%20style%3D%26quot%3Bline-height%3A%201.35%3B%20padding-left%3A%201em%3B%20text-indent%3A-1em%3B%26quot%3B%26gt%3B%5Cn%20%20%26lt%3Bdiv%20class%3D%26quot%3Bcsl-entry%26quot%3B%26gt%3BLutgen%2C%20Anne-Marie%2C%20Alistair%20Plum%2C%20and%20Christoph%20Purschke.%20%26%23x201C%3BA%20Subword%20Embedding%20Approach%20for%20Variation%20Detection%20in%20Luxembourgish%20User%20Comments.%26%23x201D%3B%20arXiv%2C%20February%2012%2C%202026.%20%26lt%3Ba%20class%3D%26%23039%3Bzp-DOIURL%26%23039%3B%20href%3D%26%23039%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2602.11795%26%23039%3B%26gt%3Bhttps%3A%5C%2F%5C%2Fdoi.org%5C%2F10.48550%5C%2FarXiv.2602.11795%26lt%3B%5C%2Fa%26gt%3B.%26lt%3B%5C%2Fdiv%26gt%3B%5Cn%26lt%3B%5C%2Fdiv%26gt%3B%22%2C%22data%22%3A%7B%22itemType%22%3A%22preprint%22%2C%22title%22%3A%22A%20Subword%20Embedding%20Approach%20for%20Variation%20Detection%20in%20Luxembourgish%20User%20Comments%22%2C%22creators%22%3A%5B%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Anne-Marie%22%2C%22lastName%22%3A%22Lutgen%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Alistair%22%2C%22lastName%22%3A%22Plum%22%7D%2C%7B%22creatorType%22%3A%22author%22%2C%22firstName%22%3A%22Christoph%22%2C%22lastName%22%3A%22Purschke%22%7D%5D%2C%22abstractNote%22%3A%22This%20paper%20presents%20an%20embedding-based%20approach%20to%20detecting%20variation%20without%20relying%20on%20prior%20normalisation%20or%20predefined%20variant%20lists.%20The%20method%20trains%20subword%20embeddings%20on%20raw%20text%20and%20groups%20related%20forms%20through%20combined%20cosine%20and%20n-gram%20similarity.%20This%20allows%20spelling%20and%20morphological%20diversity%20to%20be%20examined%20and%20analysed%20as%20linguistic%20structure%20rather%20than%20treated%20as%20noise.%20Using%20a%20large%20corpus%20of%20Luxembourgish%20user%20comments%2C%20the%20approach%20uncovers%20extensive%20lexical%20and%20orthographic%20variation%20that%20aligns%20with%20patterns%20described%20in%20dialectal%20and%20sociolinguistic%20research.%20The%20induced%20families%20capture%20systematic%20correspondences%20and%20highlight%20areas%20of%20regional%20and%20stylistic%20differentiation.%20The%20procedure%20does%20not%20strictly%20require%20manual%20annotation%2C%20but%20does%20produce%20transparent%20clusters%20that%20support%20both%20quantitative%20and%20qualitative%20analysis.%20The%20results%20demonstrate%20that%20distributional%20modelling%20can%20reveal%20meaningful%20patterns%20of%20variation%20even%20in%20%26%23039%3B%26%23039%3Bnoisy%26%23039%3B%26%23039%3B%20or%20low-resource%20settings%2C%20offering%20a%20reproducible%20methodological%20framework%20for%20studying%20language%20variety%20in%20multilingual%20and%20small-language%20contexts.%22%2C%22genre%22%3A%22%22%2C%22repository%22%3A%22arXiv%22%2C%22archiveID%22%3A%22arXiv%3A2602.11795%22%2C%22date%22%3A%222026-02-12%22%2C%22DOI%22%3A%2210.48550%5C%2FarXiv.2602.11795%22%2C%22citationKey%22%3A%22%22%2C%22url%22%3A%22http%3A%5C%2F%5C%2Farxiv.org%5C%2Fabs%5C%2F2602.11795%22%2C%22language%22%3A%22%22%2C%22collections%22%3A%5B%5D%2C%22dateModified%22%3A%222026-02-18T07%3A12%3A02Z%22%7D%7D%5D%7D

Gilles, Peter, and Frans Hinskens. “Reconstructing the Spread of Word-Final /n/ Deletion in Central Franconian Dialects. Internal and Geographical Dynamics.” Journal of Germanic Linguistics, June 1, 2026, 1–41. https://doi.org/10.1017/S1470542726100208.

Milano, Emilia, Alistair Plum, Yves Scherrer, and Christoph Purschke. “Language Ideologies in a Multilingual Society: An LLM-Based Analysis of Luxembourgish News Comments,” April 30, 2026. https://doi.org/10.48550/arXiv.2604.27661.

Lutgen, Anne-Marie, Alistair Plum, Verena Blaschke, Barbara Plank, and Christoph Purschke. “Variation Is the Norm: Embracing Sociolinguistics in NLP.” arXiv, March 25, 2026. https://doi.org/10.48550/arXiv.2603.24222.

Rehlinger, Nils. “LuxMT Technical Report.” arXiv, February 17, 2026. https://doi.org/10.48550/arXiv.2602.15506.

Lutgen, Anne-Marie, Alistair Plum, and Christoph Purschke. “A Subword Embedding Approach for Variation Detection in Luxembourgish User Comments.” arXiv, February 12, 2026. https://doi.org/10.48550/arXiv.2602.11795.