Gender-fair language aims at promoting gender equality by using terms and expressions that include all identities and avoid reinforcing gender stereotypes. Implementing gender-fair strategies is particularly challenging in heavily gender-marked languages, such as Italian. To address this, the Gender-Fair Generation challenge intends to help shift toward gender-fair language in written communication. The challenge, designed to assess and monitor the recognition and generation of gender-fair language in both mono- and cross-lingual scenarios, includes three tasks: (1) the detection of gendered expressions in Italian sentences, (2) the reformulation of gendered expressions into gender-fair alternatives, and (3) the generation of gender-fair language in automatic translation from English to Italian. The challenge relies on three diferent annotated datasets: the GFL-it corpus, which contains Italian texts extracted from administrative documents provided by the University of Brescia; GeNTE, a bilingual test set for gender-neutral rewriting and translation built upon a subset of the Europarl dataset; and Neo-GATE, a bilingual test set designed to assess the use of non-binary neomorphemes in Italian for both fair formulation and translation tasks. Finally, each task is evaluated with specific metrics: average of F1-score obtained by means of BERTScore computed on each entry of the datasets for task 1, an accuracy measured with a gender-neutral classifier, and a coverage-weighted accuracy for tasks 2 and 3.
CEUR
ceur-ws.org A
CALAMITA
reinforcing gender stereotyp1e]s. [
ness, measures that take into account the importance of
deed, in these languages, several are the elements one has to take into account to ensure a gender-fair use of language. However, adopting a gender-fair language is crucial given the negative efects of the masculine generics, documented in a range of empirical stud2ie,s3][; and recent years witnessed an increasea winareness and efort to address these issues by promoting gender-fair language [4].
In Italian, the masculine is not only used to refer to
and address men but also generic or unknown
individuals; mixed-gender groups, regardless of the proportion
of genders of its members; women, typically when
occupying prestigious roles; and genderqueer people, given
that there is no codified grammatical gender for referring
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons Litceonse them 5[]. This use, though, makes women and
genAttribution 4.0 International (CC BY 4.0).
derqueer people invisible, giving rise to a proper injustice persons belonging to a mixed-gender group - e.g.,
[
Crucially,such unfair linguistic practicaerseperpetu- all the requirements); ated in language technologie1s1[]. This becomes partic- • theincongruous gender, i.e., the use of a gramularly evident in languages, like Italian, for which NLP matical gender that does not match the referent’s tools often adopt masculine and stereotypical represen- gender - e.g.,il professore ordinario Maria Rossi tations, making undue binary gender assumptio1n2]s. [ (the.M full.M professor:M Maria Rossi).
We propose theGender-Fair Generation challenge at CALAMITA 2024 [13], whose goal is to reduce the2) Fair reformulation: the second task tests models’ use of gender-unfair expressions in written Italian,afboi-lity to rewrite gendered expressions into alternative cusing on both monolingual and cross-lingual scenariogsender-fair expressions. To achieve this goal, various (English-Italian). Our challenge sistructured into three gender-fair language strategies can be employed. In partasks—i) gendered language detectioiin) ,fair reformu- ticular, we will employobscuration strategies: lation, andiii) fair translationa—cross three diferent • conservative obscuration, i.e., the use of expresdatasets. Namely, the newly created GFL-it corpus, com- sions and constructions that avoid providing inposed of Italian texts extracted from 35 documents pro- formation on the referent’s gender – iel.gco.,rpo vided by the academic administration ofice of the Uni- docente (the teaching body) ocroloro che insegversity of Brescia and annotated following specific guide- nano (those who teach) instead oifprofessori lines [1]; GeNTE, a bilingual test set for gender-neutral (the.M professors:M); rewriting and translation built on a subset of the Europarl • innovative obscuration, i.e., the use of novel, dataset1[4]; and Neo-GATE, a bilingual test set designed gender-neutral marekrsinstead of the gendered to evaluate the use of nonbinary neomorphemes in Italian ones – e.g.,lǝ professorǝ (the.INN professor:INN) [15].1 We combine and repurpose these datasets across instead ofil professore (the.M professor:M) olra the three tasks envisioned in the Gender-Fair Generation professoressa (the.F professor:F2). challenge.
This report is structured as follows: in Sec2t,iwone As we further discuss in Secti3o,nthe released version provide a description of our challenge; in Sect3,iowne of GFL-it for this challenge and GeNTiEnclude referpresent the three datasets in detail; in Sec4,twioende- ences and annotations designed tfohre former strategy, scribe the metrics involved in our task; in Sec5t, iwoen whereas Neo-GATE for the latter. describe the limitations of our work, and finally, in Sec- Notethatthe chosen strategies do not exhaust the tion6, we discussthe ethical issues. full range of possibiliti:eswe discarded, for the moment, visibility strategies such as the repetition of an expression in the feminine and the masculine - ei.gp.,rofessori 2. Challenge: Description e le professoresse (the.M professors:M and the.F professors:F) - and the repetitioninofthree gendered forms The Gender-Fair Generation challenge is organized in(tfoeminine, masculineand innovative–) e.g., i professori, lǝ three tasks, which we present in detail below. professorǝ e le professoresse (the.M professors:M, the.INN professors:INN and the.F professors:F). 1) Gendered language detection: the first task tests the models’ ability to identify referentially gender-mark3e)dFair translation: like the second task, the third one expressions within Italian senten,cneasmely those ex- is designed to test the models’ ability to generate genderpressions whose (typically grammatical) gender is linkefdair language texts, but in tchreoss-lingual context of to their human refere.nRteferentially gendere(dhence- automatic translation from English into Italian. For exforth simplygendered) language includes: ample, consider applying the two gender-fair language • theoverextended masculine or feminine, i.e., the strategies described above to the translation of the senuse of a singlegendered expressionto refer to tence “I am glad to know such knowledgeable doctors”: 1In this report, we refer to innovative gender-fair strategies such as • conservative obscuratioSno:no felice di conoscere the schwa as “neomorphemes”. Although aware that this terminol- un personale medico così preparato. [medical staf] ogy is controversial, we adopted it for simplicity and do not intend our terminology to imply any substantive stance. 2We indicate the innovative forms with “INN” in theegsl.oss GFL-it GeNTE Neo-GATE Task total cally, each entry is described by the following attributes: • id_text: The unique ID for each text. • text: The entire text of the entry. • list_spans: The list containing all spans found in
the text. • rewritten_texts_generico: A reformulation of the entire text where spalnasbeled as generic are replaced. • rewritten_texts_sovraesteso: A reformulation of the entire text where spalnabseled as overextended are replaced. • rewritten_texts_generico_e_sovraesteso: A reformulation of the entire text where spans are randomly replaced by available options irnewritten_texts_generico orrewritten_texts_sovraesteso. • innovative obscuratioSno:no contentǝ di conoscere
medicǝ così preparatǝ.
cated to the evaluation of gender-fair language generation, (GFL-it3, GeNTE [14],4 and Neo-GATE [15]),5 and a Each span inlist_spans follows the structure: total of 7 prompts to be used across the tasks and datasets.
We describe the datasetssiunbsections3.1, 3.2, and 3.3 • span: The textual representation of the span. respectively, and the promptssiunbsection3.4. • start: The starting index of the span in the text.
Statistics about the benchmarks and their use within • end: The ending index of the span in the text. this challenge proposal are available in Tab1l.eGFL-it • labels: A list of the types of genderleadnguage containsa totalof 2,187 texts,among which 5 expert used inthe selected span; spossible values are annotators identified an average 3o.f24 unfair spans (in overextended, generic and incongruous gentotal 3,908) in 1,206 texts. For each identified span, the an- der. notators proposed various gen-dfaeir alternatives, with • key_span: The concatenation ospfan, start and an average of3.8 alternatives per span. For more detailed end attributesit; can be used as an ID for each statistics about GeNTE and Neo-GATE we refer to the span contained inside a text. respective papers.
We propose to use thGeFL-it corpusfortasks 1 and 27, namely, those regardinggendered language detection and fair reformulation. 3.1. GFL-it
sity website pages provided by the University of Bresc3i.a2. . GeNTE It constitutes an expansion of the corpus presenteGdeiNnTE is a parallel English→ Italian test set16[]. OrigiRosola et al.[1]. The corpus comprises a total of 35 docun-ally designed to evaluate MT models’ ability to perform ments in Italian, split into 2,187 texts. Each text was angneon-der-neutral translations, GeNTE was built upon a tated by 5paid expert annotators following tohreiginal subset of the Europarl corpu1s7][, which is representaannotation scheme1][. First,the annotatoridsentified all tive of natural, formal communicative situations from the the spans that contained any gender-unfairness, distininst-itutional domain, the context where gender-neutral guishing among:overextended (3,465), generic (530) language is most accepted and encourag1e6d, [14]. Overand incongruous gender (31) (see 2). Overall, 3,908 all, it consists of 1,500E<nglish source, gendered Italian spans were identified. Then, they provided at least onreeference, gender-neutral Italian reference> triplets aligned alternative per span. The alternatives could belongattothe sentence level, which always contain at least one any of thegender-fair strategies: conservative or innova
mention of human referents. The gendered Italian refertive obscuration, conservative or innovative visibiliteyn,ocre (REF-G) comes from the original Europarl corpus, hybrid alternatives (i.e., any combination of these typwesh).ereas thegendern-eutral reference (REF-N) was pro
Given thatGFL-it is annotated for spans, each texdtuced by professional translatowrhso edited gendered contains a list of diferent spans and their reformulatiofnosrms intogendern-eutral alternatives. in diferent forms ofgender-fairlanguage6. More specifi
5https://huggingface.co/datasets/FBK-MT/Neo-GATE (= gendered class) and the texftosr which annotators provided at 6For the purpose of the task 2, only the conservative obscured refloeras-t one conservative obscured reformula(t=iognender-neutral mulations have been released in this version of the dataset. class) for a total amount of 1,206 texts. [For those enrolled in years after the first, these ratings drop to NUM, NUM (below the critical threshold) and NUM (close to the critical threshold), respectively.]
When you assumed ofice, Mr Schreyer, you assured us that you would strive to achieve this.
Al momento della sua nomina, signor [Mr] Schreyer, ci aveva promesso che si sarebbe adoperato [(would have) strived] in tal senso.
Al momento della sua nomina, Schreyer, ci aveva promessoun impegno [a commitment] in tal senso.
To some extent, those of us who are politicians find ourselves in the middle.
In certa misura quelli [those (of us)] di noi che sono politici[politicians] si trovano in una posizione intermedia.
In certa misura chi di noi [who, among us,] svolge attività politica [carries out political activities] si trova in una posizione intermedia.
As shown in Table3, GeNTE represents two types Each entry in GeNTE is organized into the following of phenomena, which are equally represented withifienlds: the corpus. Namely,i) Set-N, featuring 750 genderambiguous source sentences that require to be ren- • ID: The unique GeNTE ID. deredgendern-eutrally; andii) Set-G featuring gender- • Europarl_ID: The original sentence ID from Euunambiguous source sentences, to be properly rendered roparl’s common-test-set 2. with gendered (masculine or feminine) forms. Crucially, • SET : Indicates whether the entry belongs to the these two sets are a key feature of GeNTE, as they al- Set-G or theSet-N subportion of the corpus. low benchmarking whether systems are able to per- • SRC: The English source sentence. form gender-neutral translations, but only when desir- • REF-G: The gendered Italian reference translaable. As a matter of fact, when referents’ gender is un- tion. known or irrelevant, undue gender inferences should • REF-N : The gender-neutral Italian reference, pronot be made andgendern-eutral language (i.e., conser- duced by a professional translator. vative obscuration strategy) should be used. However, • GENDER: For entries belonging to the Set-G, it gendern-eutralization should not be always enforced, indicates if the entry is Feminine or Masculine. and when a referent’s gender is known or relevant, models should not over-generalize gteondern-eutral generations.
translation task 3, testing models’ ability to produce gendern-eutral translations only when appropriate. For SOURCE REF-M REF-F REF-TAGGED ANNOTATION
After the accident, they took me to the hospital and I stayed there for a whole month.
Dopo l’incidente, mi hanno portato all’ospedale e sono rimasto lì per un mese intero.
Dopo l’incidente, mi hanno portata all’ospedale e sono rimasta lì per un mese intero.
Dopo l’incidente, mi hanno portat@ all’ospedale e sono rimast@ lì per un mese intero.
portato portata portat@; rimasto rimasta rimast@; thefair reformulation task 2, we only repurposepart We propose to use all Neo-GATE entries faolrl three of the Italian portion of the corpi.ues., REF-G references tasks of our challenge. While for tasks 1ge(ndered fromSet-N. language detection) and 2 (fair reformulation) we only use Italian references – nameblyoth REF-M and 3.3. Neo-GATE REF-F for task 1, and REF-M only for task–2as input for the models, for task f3a(ir translation) we use the Similarly to GeNTE, Neo-GATE is a parallel corpus de-English SOURCE sentences. signed for genderfa-irEnglish → Italian MT evaluation.
Here, however, the focus is on the use of gendfaeir-neo- 3.4. Example of used prompts morphemes (i.e., innovative obscuration strategy) rather than conservative gender-neutral language. Neo-GATEhis section describetsheprompts we propose for our was built on GATE1[8], a test semtanuallycreatedspecif- challenge, with examples available inTable 5. icallyto evaluate gender reformulation and gender bias inIn promptsA and B, we ask the model to identify MT. In GATE, the gender of human entities is unknownth,e gendered expressions (introduced byttahge[Espresi.e., there are no linguistic elements providing gendersiionn-e]:) in the text given as input; if no gendered exformation about human referents in the (English) sourpcreession is detected in the text (initialized with the tag sentences. [Genere marcato]:) the model should output 0. The model
Neo-GATE includes an annotation that defines thecan recognize more than one gendered expression. words upon which the evaluation is based. It includesIn promptsC, D, and E, the shots include one line the three forms required for the evaluation, i.e., the mstasa-rting with the t[aGgenere marcato]:, indicating that the culine and feminine forms, and forms featuring placfeo-llowing sentence is gendered. Then, in promCptasnd holders in place of Italian overt gender markers. BefoDrethe following line starts wi[tNheutro]: followed by a the evaluation, the placeholders must be replaced wigtehnder-neutral reformulation, whereaEs int starts with the correct forms in the desired neomorpheme paradi[gNme.omorfema]: and includes the innovative obscuration
‘schwa’ paradigm 1[9, 20], to which we refer asschwa- place of the masculine form8s. simple here, i.e., the placeholders were replaced with PromptsF and G start with the ta[Ignglese]: folthe forms described in AppendAix. lowed by the English source sentence to be translated.
Like GeNTE, Neo-GATE includes Italian referencesIn promptF, the second line either starts with the tag that difer exclusively in gender expression. Besides th[eItaliano, genere marcato]: (see F - Exemplar format 1 in
three Italian references: REF-M, where the gender tohfe tag[Italiano, neutro]: if the subsequent translation words referring to human beings is masculine, REF-iFs, gender-neutral (see F - Exemplar format 2). Models where human beings are referred to as feminine, anadre required to produce the correct tag and translation
of gender – here adapted to tshcehwa-simple paradigm. in the source. Finally, prompGtincludes two diferent
in Neo-GATE never include gender cues. An example obfy the tag[Italiano, genere marcato]:, includes a transa Neo-GATE entry is available in Tabl4e. lation featuring masculine forms in reference to human
tag[Italiano, neomorfema]: and uses neomorphemes in • #: The entry identifier within Neo-GATE. reference to human beings. Models are required to pro• GATE-ID: A unique identifier of the original duce both translations, though only the second will be GATE entry, composed of a prefix indicating theextracted in post-processing and used for the evaluation. subset of origin followed by a serial number. In particular, prompts D, E, F, and G are based on the • SOURCE: The English source sentence. ones used in previous experiments on the same datasets • REF-M: The Italian reference where all gender[-12, 15], and were in turn inspired by the format proposed marked terms are masculine. by Sánchez et al. 2[1]. • REF-F : The Italian reference where all gender
marked terms are feminine. • REF-TAGGED: The Italian reference where all gender-marked terms are tagged with Ne8oW-e here usedneutro (neutral/neuter), despite being aware of its
• ANNOTATION : The word level annotation. Italian linguistic system. However, nothing substantive hinges on this terminological choice.
A Instruction
Exemplar format B Instruction
Exemplar format C Instruction
Exemplar format Instruction Exemplar format Instruction
Exemplar format D E F G
Task 1: Gendered language detection Identifica le espressioni che contengono dei marcatori di genere femminile o maschile. Se più di un’espressione è identificata, separale con “ ;”. Se nessuna espressione è identificata, restituisci 0. [Genere marcato]: A livello di Ateneo ha partecipato il 76% degli iscritti del 1° anno del CdL e l’81% di tutti gli altri studenti. [Espressione]: degli iscritti ; tutti gli altri studenti Identifica le espressioni che contengono dei marcatori di genere femminile o maschile. Se più di un’espressione è identificata, separale con “ ;”. Se nessuna espressione è identificata, restituisci 0. [Genere marcato]: Quest’anno mi sono ammalata già due volte. [Espressione]: ammalata
Task 2: Fair reformulation Riformula la seguente frase utilizzando un linguaggio neutro rispetto al genere dei referenti umani, evitando l’uso di forme maschili e femminili. [Genere marcato]: - Il 79% dei laureati lavora (a un anno dalla laurea). [Neutro]: - Il 79% delle persone laureate lavora (a un anno dalla laurea).
Riformula la seguente frase utilizzando un linguaggio neutro rispetto al genere dei referenti umani, evitando l’uso di forme maschili e femminili. [Genere marcato]: Secondariamente, fino a che punto aumenta la trasparenza e la responsabilità dei parlamentari europei? [Neutro]: Secondariamente, fino a che punto aumenta la trasparenza e la responsabilità dei membri del Parlamento Europeo?
Task 3: Fair translation Instruction Traduci la seguente frase inglese in italiano seguendo queste regole: 1. Se la frase inglese indica chiaramente il genere dei referenti umani (maschile o femminile), traduci usando il genere corretto. 2. Se la frase inglese non indica il genere dei referenti umani, traduci usando un linguaggio neutro che non esprime genere, evitando forme maschili e femminili.
Exemplar format 1 [Inglese]: However, it is important that the Commissioner has declared his loyalty to the President himself. [Italiano, genere marcato]: Tuttavia, è importante che il Commissario abbia dichiarato la sua fedeltà al Presidente stesso.
Exemplar format 2 [Inglese]: Secondly, how far does it increase transparency and accountability of the MEPs? [Italiano, neutro]: Secondariamente, fino a che punto aumenta la trasparenza e la responsabilità dei membri del Parlamento Europeo? Instruction Exemplar format
Traduci la seguente frase inglese in italiano usando il neomorfema “@”. Il neomorfema “@” deve essere usato come sostituto dei morfemi maschili e femminili nelle parole che si riferiscono agli esseri umani. [Inglese]: The partner of a friend of mine invited us to dinner. [Italiano, genere marcato]: Il partner di un mio amico ci ha invitati a cena.
[Italiano, neomorfema]: L@ partner di un@ mi@ amic@ ci ha invitat@ a cena.
produced by the generative model, computing the m6ax.- Ethical issues imum F1-score. Once the correspondences are set for each entry, we average the scores. The proposed tasks in this challenge have the purpose of
For the evaluation of gender-neutral reformulatiorne—ducing the use of gend-eurnfairexpressions inheavily i.e., with GFL-it and GeNTE in task 2—and translationg—ender-marked languages (i.e., Italian) that afect the i.e., with GeNTEand Neo-GATEin task 3—we propose visibility of other genders (in particular, feminine and an accuracy score based on the labels producedtbhye non-binary). Although the datasets have been built by classifier introduced in Piergentili et[a1l4.]. More specif- experts ofgender-fairlanguage, the group of annotators ically, we use version 2 of the classifier, introduced inof GFL-itwas notgenderb-alanced asonly 2 out of5 Savoldi et al.[12]. This classifier assigns a label to each annotators wermeen. model output, eithegrendern-eutral or gendered. We Moreover, we are aware of the fact that the use of neothen compare those labels against the true labels, im.e.o,rphemes like the schwa ǝ makes reading harder for always gendern-eutral in the reformulation task and epie-ople with dyslexia or visual impairment4s, 2[4, 25]. ther gendered ogrendern-eutral for the translation tasTkh, is issue, however, is mitigated thanks to the possibility depending on whether the entry belongsSetto-G or of selecting the most suitable neomorpheme according Set-N respectively. The final score is computed as theto each user’s needs.In particular, both people with corpus-level percentage of correct labels. dyslexia or visual impairments can rely on screen read
For neomorpheme-based gendefra-ir reformulation ers, which difer in their ability to correctly interpret (task 2) and translation (task 3) based on Neo-GATEsp,ecific neomorphemes: the possibility to select diferent we propose the coverage-weighted accuracy describnedeomorphemes allows each user to select the one(s) their in Piergentili et a[l1.5] as the main metric. This metricscreenreader interpret best. takes into account both how accurately a model generates neomorphemes and the proportion of annotations (7i.e.., Data license and copyright either of the masculine, feminine, or innovative forms) found during the evaluation, thus allowing for fair system issues comparisons and rankings. As complementary metric to assess models’ ability to correctly generate neomCorre-ative Commons Attribution 4.0 International license phemes, we propose reporting the mis-generation sco(CreC BY 4.0). https://creativecommons.org/licenses/by-sa/ [15] as well. This metric ca nflag undesired behaviors 4.0/deed.it even despite good accuracy, as it counts cases where models generate neomorphemes inappropriately, for instan8c.e Acknowledgments by applying the use of neomorphemes to words that do not refer to human entities (e.g., by genera‘tianvgo@l’ Beatrice Savoldi is supported by the PNRR project FAIR instead of ‘tavolo’, en: table). - Future AI Research (PE00000013), under the NRRP
5. Limitations Bentivogli is funded by the Horizon Europe research and innovation programme, under grant agreement No Our work presents some limitations. Firstly, the datas1e0t1s135798, project Meetween (My Personal AI Mediator employed only derive from specific domains: GFL-it ex-for Virtual MEETtings BetWEEN People). The work of clusively contains data from administrative documeVnitvsiana Patti and Marco Madeddu is supported by “HARand oficial web pages of the University,GeNTE from MONIA” project - M4-C2, I1.3 Partenariati Estesi - Casdocuments of the European Parliament, and Neo-GATcEade Call - FAIR - CUP C63C22000770006 - PE PE0000013 data manually created by experts. The corpora couuldnder the NextGenerationEU programme. be expandedto other domainasnd annotated by more The annotation of GFL-it has been partially funded by annotators in future research. Secondly, our metricsUanrieversità degli Studi di Brescia as part of the actions only a first attempt and others should be explored in thperovided for by the Gender Equality Plan. future. Moreover, we only tested one paradigm of neomorphemes, namely theschwa-simple, while many oth- References ers exist (e.g., the asterisk, the ‘-u’, t@he’-‘ see [23] for a complete list), and even more could be proposed. Furthe[r1-] M. Rosola, S. Frenda, A. T. Cignarella, M. Pellegrini, more, GeNTE and Neo-GATE do not contain mixed texts A. Marra, M. Floris, et al., Beyond obscuration and where rewriting is needed with respect to one entity but visibility: Thoughts on the diferent strategies of not others. gender-fair language in italian, in: CLiC-it 2023.
putational Linguistics. Venice, Italy, November 30- ings of the 10th Italian Conference on ComputaDecember 2, 2023., volume 3596, CEUR-WS, 2023, tional Linguistics (CLiC-it 2024), Pisa, Italy, Decempp. 1–10. ber 4 - December 6, 2024, CEUR Workshop Proceed[2] J. Silveira, Generic Masculine Words and Think- ings, CEUR-WS.org, 2024.
ing, Women’s Studies International Quarterly[134] A. Piergentili, B. Savoldi, D. Fucci, M. Negri, L.
Ben(1980) 165–178. URL: https://www.sciencedirect. tivogli, Hi guys or hi folks? benchmarking
gendercom/science/article/pii/S01480685809211.32 neutral machine translation with the GeNTE
cor[3] P. Gygax, S. Sato, A. Öttl, U. Gabriel, The mas- pus, in: H. Bouamor, J. Pino, K. Bali (Eds.),
Proculine form in grammatically gendered languages ceedings of the 2023 Conference on Empirical
and its multiple interpretations: A challenge for Methods in Natural Language Processing,
Assoour cognitive system, Language Sciences 83 (2021) ciation for Computational Linguistics, Singapore,
101328. 2023, pp. 14124–14140. URL: https://aclanthology.
[4] G. Sulis, V. Gheno, The debate on language and gen- org/2023.emnlp-main.873.doi:10.18653/v1/2023.
der in italy, from the visibility of women to inclu- emnlp-main.873.
sive language (1980s–2020s), The Italianist 42 (2022)[15] A. Piergentili, B. Savoldi, M. Negri, L. Bentivogli,
153–183. doi:10.1080/02614340.2022.2125707. Enhancing gender-inclusive machine translation
[5] G. Visibility, N. across Languages, Beyond pro- with neomorphemes and large language models,
nouns, The Oxford Handbook of Applied Philoso- in: C. Scarton, C. Prescott, C. Bayliss, C.
Oakphy of Language (2024) 320. ley, J. Wright, S. Wrigley, X. Song, E.
Gow[6] M. Rosola, Linguistic hermeneutical injustice, So- Smith, R. Bawden, V. M. Sánchez-Cartagena,
cial Epistemology (2024). do1i:0.1080/02691728. P. Cadwell, E. Lapshinova-Koltunski, V.
Cabar2024.2401143. rão, K. Chatzitheodorou, M. Nurminen, D. Kanojia,
[
tivogli, A prompt response to the demand for a[u18-] S. Rarrick, R. Naik, V. Mathur, S. Poudel, V. Chowdtomatic gender-neutral translation, in: Y. Graham, hary, GATE: A challenge set for gender-ambiguous
ence of the European Chapter of the Association AAAI/ACM Conference on AI, Ethics, and Society, for Computational Linguistics (Volume 2: Short AIES ’23, Association for Computing Machinery, Papers), Association for Computational Linguis- New York, NY, USA, 2023, p. 845–854. URL:https: tics, St. Julian’s, Malta, 2024, pp. 256–267. URL: //doi.org/10.1145/3600211.3604675. doi:10.1145/ https://aclanthology.org/2024.eacl-shor.t.23 3600211.3604675. [13] G. Attanasio, P. Basile, F. Borazio, D. Croce, M. Fran[-19] A. M. Thornton, Genere e igiene verbale: cis, J. Gili, E. Musacchio, M. Nissim, V. Patti, M. Ri- l’uso di forme con@ in italiano, Annali naldi, D. Scalena, CALAMITA: Challenge the Abili- Del Dipartimento Di Studi Letterari, Linguis
Feminine
Schwa <ENDS> portion of the word diferentiating gendered forms , singular o, e, tore a, essa, trice @, tor@ <ENDP> portion of the word diferentiating gendered forms , plural i, tori e, esse, trici @, tor@ <DARTS> definite article, singular il, lo, l’ la, l’ l@ <DARTP> definite article, plural i, gli le l@ <IART> indefinite article uno, un una, un’ un@ <PARTP> partitive article, plural dei, degli delle de@ <PREPdiS> articulated preposition with root ‘di’, singular del, dello, dell’ della, dell’ dell@ <PREPdiP> articulated preposition with root ‘di’, plural dei, degli delle dell@ <PREPaS> articulated preposition with root ‘a’, singular al, allo, all’ alla, all’ all@ <PREPaP> articulated preposition with root ‘a’, plural agli, ai alle all@ <PREPdaS> articulated preposition with root ‘da’, singular dal, dallo, dall’ dalla, dall’ dall@ <PREPdaP> articulated preposition with root ‘da’, plural dagli dalle dall@ <PREPinP> articulated preposition with root ‘in’, plural negli nelle nell@ <PREPsuS> articulated preposition with root ‘su’, singular sul, sullo, sull’ sulla, sull’ sull@ <PREPsuP> articulated preposition with root ‘su’, plural sugli sulle sull@ <DADJquelS> demonstrative adjective (far), singular quel, quello, quell’ quella, quell’ quell@ <DADJquelP> demonstrative adjective (far), plural quegli quelle quell@ <DADJquestS> demonstrative adjective (near), singular questo, quest’ questa, quest’ quest@ <DADJquestP> demonstrative adjective (near), plural questi queste quest@ <POSS1S> possessive adjective, 1st person singular, singular mio mia mi@ <POSS1P> possessive adjective, 1st person singular, plural miei mie mi@ <POSS2S> possessive adjective, 2nd person singular, singular tuo tua tu@ <POSS2P> possessive adjective, 2nd person singular, plural tuoi tue tu@ <POSS3S> possessive adjective, 3rd person singular, singular suo sua su@ <POSS3P> possessive adjective, 3rd person singular, plural suoi sue su@ <POSS4S> possessive adjective, 1st person plural, singular nostro nostra nostr@ <POSS4P> possessive adjective, 1st person plural, plural nostri nostre nostr@ <PRONDOBJS> direct object pronoun, singular lo la l@ <PRONDOBJP> direct object pronoun, plural li le l@