Morphological vs. Lexical Antonyms in Italian: A Computational Study on Lexical Competition Martina Saccomando1,∗ , Andrea Zaninello2 and Francesca Masini1 1 Alma Mater Studiorum - University of Bologna, Bologna (Italia) 2 Fondazione Bruno Kessler, Trento; Free University of Bozen-Bolzano (Italy) Abstract In this paper, we examine the competition between pairs of adjectives in Italian that are antonyms of the same term: one is a “morphological antonym” formed by negative prefixation, the other is a “lexical antonym” with no morphological relationship with the term in question. We consider pairs of adjectives that are reported as antonyms in lexicographic resources and extract the nouns that can be modified by both adjectives from a large corpus. We select a set of 8 nouns for each pair that present higher, lower, and comparable frequencies combined with each antonym respectively and then we perform two experiments with a LLM. Firstly, we perform experiments for masked-token prediction of the adjective, to study the correlation between prediction accuracy and the frequency of the noun-antonym pair. Secondly, we perform a polarity-flip experiment with a multilingual LLM, asking to change the adjective into its positive counterpart. Our results point to the conclusion that the lexical antonym seems to have a narrower lexical coverage and scope than the morphological antonym. Keywords Antonymy, Morphological antonyms, Lexical antonyms, Competition, Corpus analysis, Large language model, Token prediction, Polarity flip, Italian 1. Introduction felice infelice Antonymy is the semantic relationship between terms triste with opposite meanings. In their canonical form, two antonyms’ meanings can be represented as the poles of a semantic continuum [1] where one term has a “positive” Figure 1: Two possible antonyms, one morphological, one semantic value, the other a “negative” one [2]. lexical, for the same word. In Italian, given a word (e.g., the adjective felice ‘happy’), antonyms can either be realized via prefixa- tion of that word (e.g., infelice ‘unhappy’) or through lexical antonyms; explicitly in the case of morphological an independent lexeme (e.g., triste ‘sad’). In our work, antonyms, by adding a prefix with a negative, contra- we refer to these types of antonyms as morphological dicting value. Considering their different morphological antonym and lexical antonym, respectively. A word in structure, one possible hypothesis on their lexical com- the lexicon may have both a morphological and a lexical petition is that the morphological antonym has a more antonym, only one of them, or neither. In this paper, we restricted semantics, representing the negation of the se- are interested in triplets of adjectives where a positive mantics of its adjectival base, while the lexical antonym adjective (e.g., felice) presents two possible antonyms (or has a broader semantic coverage, as it is morphologically “co-antonyms”), one formed morphologically by prefix- independent from its positive counterpart (see Section ation (e.g., infelice), one morphologically independent 3). (e.g., triste) (Figure 1). To the best of our knowledge, despite the wealth of In this paper, we are interested in studying the factors literature on antonyms (Section 2), there is no empirical that govern the selection of the morphological antonym in-depth study that investigates the competition between vs. the lexical one. These two types of antonyms express morphological and lexical antonyms in single languages, “negative” semantics with respect to the opposite, “pos- including Italian. Studies on antonyms do identify the itive” term in different ways: implicitly in the case of two types of antonyms but generally do not address the factors influencing the preference for one type over the CLiC-it 2024: Tenth Italian Conference on Computational Linguistics, other intralinguistically. Dec 04 — 06, 2024, Pisa, Italy This study investigates the competition between these ∗ Corresponding author. two types of antonyms by firstly studying their distri- Envelope-Open martina.saccomando@studio.unibo.it (M. Saccomando); bution in corpora (Section 5.1); secondly, by testing the azaninello@fbk.eu (A. Zaninello); francesca.masini@unibo.it (F. Masini) ability of a native-Italian language model to predict them © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License in a masked-token prediction task (Section 5.2); and, fi- Attribution 4.0 International (CC BY 4.0). CEUR ceur-ws.org Workshop ISSN 1613-0073 Proceedings nally, by performing a substitutability task within the only one of the antonyms possesses, preventing their full same context by switching the polarity of the context sen- synonymy and interchangeability. tence with a SOTA multilingual LLM, in order to study Justeson and Katz [12] take a different approach. Using when the adjective is switched to the positive un-prefixed the Brown Corpus and Deese’s antonym dataset, they adjective or to another, positive but morphologically un- were among the first to study antonymy based on a cor- related one (Section 5.3). pus. They found that antonymic terms co-occur more frequently than expected, confirming a syntagmatic rela- tionship between them (in addition to the paradigmatic 2. Background one). This syntagmatic relationship was confirmed by a more extensive work carried out in subsequent years: Whereas the exploration of the competition between mor- Jones [11] collected 56 antonym pairs, analyzing jour- phological and lexical antonyms, addressed in this paper, nalistic texts, to identify eight discursive functions of has not gained much attention so far, the literature on antonym co-occurrence. antonymy in general is abundant, especially in relation A study that does address the competition between to the English language. forms is Aina et al. [13], who studied syntactically A term’s antonym is related to it according to three negated adjectives and morphological antonyms (e.g., main characteristics: polarity, gradability and canon- not happy vs unhappy, respectively). Using distributional icity. The first two characteristics refer to the position- semantics they found that a syntactically negated ad- ing of the two antonymic terms w.r.t. the two poles jective is more similar to the positive adjective than to (polarity) of a graded (gradability) scale [3], along its lexical antonym. Additionally, they show that the which free positions can be occupied by other similar morphological antonym is less similar to its lexical base but differently graded terms. The scale is based on a spe- compared to the similarity between a negated adjective cific property that the two terms share. For example, in and its non-negated counterpart. the pair long-short, the two antonyms share the property Last but not least, a very recent typological study [14] of “length”, defining the start and end of an axis whose examines 37 antonym pairs across 55 languages, focusing poles are defined by two terms, with long representing on antonym formation. When a derived form is attested, the “unmarked” base term of the opposition (this is why it typically applies to the member of the pair with lower we ask for “how long” rather than “how short” something valence or lesser magnitude. Antonyms related to core would last [4]). property concepts (dimension, age, value, color) tend to However, there are cases where antonymic pairs be expressed through distinct lexical forms (resulting in are formed with potentially competiting antonyms, lexical antonyms), while those related to peripheral prop- like friendly-unfrendly vs. friendly-hostile: friendly- erty concepts (physical property, human property, speed) unfriendly is placed on a scale that defines a greater or are generally encoded using derived forms (namely mor- lesser degree (gradability) of a property, while friendly- phological antonyms). Although the study is insightful hostile are certainly in opposition but belong to two scales and inspiring in a number of ways, the specific question of incompatible properties. In terms of their gradability, of which reasons underlie the preference for lexical or therefore, it seems that the morphological antonym is morphological forms in a single language remains unan- “more gradable” than the lexical one. swered. Canonicity, according to Paradis and Willners [5], defines two semantically related and conventionalized terms as a pair in the language. It is a gradual property 3. Morphological vs. Lexical and can be possessed to a greater or lesser extent. A high degree of canonicity translates into a high degree Antonyms of semantic-lexical embedding in memory and leads to A morphological antonym (e.g., inattivo ‘inactive’ conventionalization in usage. from attivo ‘active’) is immediately recognizable as a Psycholinguistic studies also suggest that canonical negative term due to the presence of the negative prefix antonyms derive from the speakers’ experience with the un- ‘in-/un-’, while the nature of the opposition is less language: the two terms are inseparable, one elicits the immediate with a lexical antonym (e.g., passivo ‘pas- other [6, 7, 8, 9, 10]. When a term has two structurally sive’), because one has to identify the property shared by different but semantically similar antonyms (Figure 1), the two opposing terms. canonicity is influenced by factors such as learned pref- Moreover, these two ways of forming an antonym erence for specific pairings, the speaker’s familiarity due from the same term create an asymmetric system: while to exposure, and different nuances of meaning [11]. Out one of the two terms in lexical opposition (e.g., attivo of context, the antonyms may appear equivalent, but ‘active’) has its morphological antonym (e.g., inattivo ‘in- within context, a specific meaning may be activated that active’), the other term (e.g., passivo ‘passive’) does not a wider range of ways of not being friendly (such as being (e.g., *apassivo, *impassivo, ‘*unpassive’). This imbalance aloof) whereas hostile is fairly specific” (Murphy 2003: is due to the greater emphasis that language places on 202). So, the morphological antonym would be more everything that requires more precise specifications [15]. polysemous, while the lexical counterpart would have The situation is not always perfectly equivalent interlin- a narrower scope. guistically: for example, a form like *invero to indicate Given these two competing hypotheses, we aim to the opposite of vero ‘true’ is not attested in Italian, while empirically verify: it is possible in English (untrue). However, there are cases in both Italian and English • whether the lexical antonym is more frequent where the two competitors have different nuances: for than the morphological antonym; example, infelice-triste / unhappy-sad cover different con- • whether the morphological antonym is actually texts in that triste and sad convey a stronger emotional less polysemous than the lexical antonym. meaning, while infelice and unhappy encompass certainly In order to do this, we design a set of experiments. We strong but less intense feelings.[16] first select antonym pairs; we calculate their frequency Regarding distribution in usage, lexical antonyms of co-occurrence with nouns to have a defined context; are predominant for more basic meanings, supporting the then, we perform two tasks: (i) masked-token prediction Principle of Least Effort theorized by Zipf [17], which sug- and (ii) polarity flip. gests that we expect the most used concepts to be coded with short and simple words: basic terms therefore tend to have structurally simple antonyms even when more 4. Dataset Construction complex morphological antonyms would be possible. For example, for a pair like alto-basso ‘tall-short’, there is no 4.1. Antonym Pair Selection morphological counterpart that can be associated with ei- ther term: neither *inalto ‘*untall’ nor *inbasso ‘*unshort’ For our study, we decided to focus on adjectives, as this exist. These are canonical antonyms referring to basic class is the most suitable for investigating antonymy, language concepts: in cases like this, but not exclusively, given that it includes content words that normally ex- preference for simpler and more immediate words blocks press qualities. Moreover, adjectives are semantically the potential formation of a morphological antonym. simpler compared to other word classes as they usually According to Murphy [4], culturally salient concepts describe a single property that may be or may not be necessitate clear and concise linguistic expressions. For present to a greater or lesser degree ( Jones et al. [1]). this reason, lexical antonyms (e.g., passivo ‘passive’) are Starting from 1535 adjectives of the fundamental Ital- the most frequent choice because they require less cogni- ian lexicon extracted from the Italian dictionary Zingarelli tive effort to understand. Although it is possible to create 2024, we filtered 303 items marked as ‘contr.’. new opposite terms through derivational and morpholog- We then selected, for this pilot study, 5 adjectives with ical processes with speaker’s creativity, this option is less the following properties: they needed to be adjectives commonly employed in this context, as it is perceived as with both a morphological and a lexical antonym; they a deviation from the linguistic norm. needed to be maximally interchangeable in different con- The competition between the two terms of the texts.1 antonym pair, i.e., the situation in which the usage con- Finally, we created our initial dataset by pairing each text of both terms is nearly the same and allowing for a adjective with its corresponding morphological antonym certain degree of substitutability, is still debated. and a possible lexical antonym. According to one hypothesis, since the morphological The morphological antonym was formed by using one antonym is the “perfect” negation of a specific lexical of the three possible prefixes productively used in Italian base, it should occur in more restricted contexts (i.e., a to create antonyms, namely: dis-, s-, in- [20]2 . subset of the contexts of its positive counterpart) and The lexical antonym was chosen randomly among all should therefore have a narrower semantics (cf. [18, possible options, taking into account synonymy with the 19, 15]). So, morphological antonyms would be less morphological antonym and semantic neutrality. This polysemous. On the other hand, the lexical antonym, means that the lexical antonyms were selected to be ide- not sharing identical lexical properties with the opposed ally substitutable with the morphological ones in as many term, should occur in broader contexts and thus be more contexts as possible, and roughly possessing the same polysemous. 1 The admittedly limited size of the dataset is due to the exploratory However, Murphy [4], examining the English triple nature of our study. friendly-unfriendly-hostile, notes that “The two antonyms 2 The prefix in- is the most productive and the most widespread. It are hardly equivalent, though, since unfriendly describes often adapts phonetically to the bases it attaches to, forming the allomorphs im-, ir-, and il- via assimilation [20]. number of senses according to the dictionary (cf. Table 2). 4.3. Lexical Context Definition and The synonymy between the two types of antonyms was Example-Sentence Extraction further confirmed using Il grande dizionario dei Sinonimi e dei Contrari, Zingarelli 2013 [21]. Subsequently, for each antonym pair, eight nouns with Summing up, the antonym pairs examined are: different co-occurrence frequencies were selected. Specif- ically, we considered both nouns that typically occur with • infelice ‘unhappy’ - triste ‘sad’ one of the antonyms (e.g., matrimonio infelice), falling • impreciso ‘imprecise’ - approssimativo ‘approximative’ within the domain of collocations, and (more generic) • scorretto ‘incorrect’- sbagliato ‘wrong’ nouns whose co-occurrence frequencies with the lexical • imprudente ‘imprudent’ - avventato ‘reckless’ and the morphological antonym are very similar (e.g., • insufficiente ‘insufficient’ - scarso ‘scant’ donna infelice and donna triste) (cf. Table 1). resulting in 5 triplets (base, morph_ant, lex_ant). Noun + Adjective Frequency matrimonio infelice ‘unhappy marriage’ 886 4.2. Corpus-based Analysis matrimonio triste ‘sad marriage’ 24 donna infelice ‘unhappy woman’ 325 We analyzed the occurrences of the selected adjectives donna triste ‘sad marriage’ 316 with nouns in the itTenTen20 corpus, a large web corpus Table 1 of written Italian, searched through the SketchEngine Differences between high and low frequency name+adjective platform https://www.sketchengine.eu/. co-occurrences The analysis of the occurrences highlighted that the two types of antonyms display partially different collo- cational preferences (see Appendix A, Table 3). The latter case is especially interesting for our current Overall, we can split the antonymic adjective-noun purposes, as it represents possible ground for “compe- couples in two groups according to whether the co- tition”, namely a situation where the context of use is occurrence is: nearly the same and allows for a certain substitutability between the two terms of the antonym pair.4 • (i) polarization towards one of the two ad- After defining the noun list, for each noun we ran- jectives: in these cases, we can speak of fairly domly selected 10 sentences containing the noun fol- stable distributional preferences, falling within lowed by the morphological antonym and ten sentences the realm of collocations or idiomaticity (e.g., mat- containing the same noun followed by the corresponding rimonio preferably selecting infelice rather than lexical antonym from itTenTen20. This was done for all triste); eight nouns and for all five antonym pairs, resulting in a • (ii) similar with the two adjectives, indicating 800-sentence dataset. potential substitutability of the two antonyms in the same contexts (e.g., donna selecting both infe- lice and triste with similar relative frequencies). 5. Experiments Both groups are relevant to explore the context of use In Section 3 we outlined two possible hypotheses re- of the two types of antonyms, although, for our current garding the competition between the two terms of the purposes, we specifically targeted the nouns in group antonym pair and we selected the following as a working (ii), namely nouns that occur with both adjectives, sug- hypothesis: the morphological antonym, being formed gesting a certain degree of competition between the two by a negative prefix applied to a specific lexical base, antonyms: see sentence 1, where infelice ‘unhappy’ can would have more restricted usage contexts (possibly a be replaced by the lexical antonym triste ‘sad’. subset of the contexts of the lexical base), and therefore be less polysemous than the lexical one; on the other 1. Un ritratto preciso ma discontinuo che ci restitu- hand, the lexical antonym, not sharing morphological isce l’immagine di una donna infelice, umiliata, structure with the opposing term, would semantically affranta, ma non distrutta, non arresa alla sorte 3 cover some or all of its meanings along with other inde- pendent meanings, resulting in broader usage contexts and greater polysemy. To verify this hypothesis, we performed 2 sets of exper- iments: (i) masked-token prediction, to estimate the 3 ‘A precise but fragmented portrait that gives us the image of an un- 4 happy, humiliated, distraught woman, but not destroyed, not resigned For a detailed view of the selected adjectives and nouns, as well as to fate’. their co-occurrence frequencies, see Appendix A. probability of occurrence of one antonym or the other ac- We used bert-base-italian-xxl-cased language model5 cording to a native Italian language model; (ii) polarity to perform a token prediction task by masking the flip, to transform the collected sentences from a negative antonym present in each sentence. The model was asked meaning to a positive meaning. to predict the probability of occurrence between the two possible antonyms; then, we took the alternative with the 5.1. Word Senses and Lexical Variety highest probability according to the model as the model prediction. Our analysis started with the identification of adjectives and their possible antonyms, which, as mentioned (Sec- 5.3. Polarity-flip tion 4.1), have been chosen on the basis of their possible substitutability within the same context. For this reason, In this task, we asked a SOTA LLM, GPT-4o, to transform first of all, the various dictionary definitions of antonyms the sentences extracted from the dataset, both those con- were taken into consideration. We counted how many taining the morphological antonym and those containing senses are associated to each antonym in the Zingarelli the lexical antonym, into positive sentences. dictionary [22], taking the number of senses reported as We used the same prompt for all antonym pairs, a first proxy of semantic broadness. parametrising the antonyms and sentences presented, As a second proxy, the semantic coverage of each by asking the model to flip the sentence from a negative antonym was taken into account. We conducted an anal- sense to a positive one, always by changing the adjective ysis of the lexical variety of each group’s context in the accompanying the target noun.6 selected sentences, by calculating the token/type ratio We then fetched the new adjective generated by the for each group. We report the results in Table 4. As can model, and calculated when the new, positive adjective be seen, no relevant differences were found according to coincided with the lexical base, and when not. these features. The rationale behind this is that the senses of an antonym can be separated through the various positive 5.2. Masked-token Prediction terms with which it can be changed. We expected that sentences containing the morphological antonym would According to our hypothesis, in this task we expect that be turned into positive using their lexical base more of- the predicted antonym will have a higher prediction ac- ten than their lexical counterparts, indicating a narrower curacy in the sentences with the highest occurrence of semantics. the adjective with the selected words (represented by high relative frequency). In contexts with similar fre- quencies, we expect that accuracy should be similar for 6. Analysis and Results both antonyms, showing a genuine competition between the two, as the language model should not have a specific As regards the number of meanings listed in the dictio- preference. nary for the two terms of the antonym pair (Table 2), We previously said that the words that form the these are almost the same, indicating that the recognized antonym pairs can be considered synonymous. In fact, senses of each antonym alone are not decisive to de- full synonyms are rather rare (Murphy [4], among oth- termine the selection between one or the other. As for ers), also because languages tend to avoid synonymy by lexical variability (see Appendix B, Table 4), token/type differentiation in terms of meaning or distribution [23]. ratio also fails to reveal a significant difference between Therefore, the two terms of the pair are better regarded as morphological and lexical antonyms; in only three cases near-synonyms, meaning that one term can cover almost does the token/type ratio of lexical antonyms exceed all the meanings of the other but not all of them. that of morphological antonyms. This seems to indicate To evaluate the factors that lead to the choice of one that factors other than contextual variability underlie the antonym over the other, we decided to observe how a na- preference for one or the other. tive Italian language model pre-trained for masked-token Let us now consider Table 3, Appendix A. In addition to prediction model behaves in terms of the probability of the co-occurrences of nouns with the two antonyms, the occurrence of an antonym in a given context. In this accuracy for the two tasks for each noun-adjective pair is respect, see Niwa et al.[24], who used BERT to predict also provided (further divided into noun-morphological antonyms in specific contexts: experiments on Japanese antonym and noun-lexical antonym). The distribution slogans showed a top-1 accuracy of 29.3% and a top-10 of nouns in this table follows a specific order: nouns at accuracy of 53.8%, with human evaluations confirming the extremes exhibit occurrence frequencies polarized that over 85% of predicted antonyms were appropriate, 5 dbmdz/bert-base-italian-xxl-uncased demonstrating the method’s effectiveness in capturing 6 For detailed information on the prompt used in this study, please contextually relevant antonyms. refer to the Appendix B. accuracy accuracy polarity polarity senses senses rel.freq rel.freq mask token mask token flip flip prediction prediction Couple Exp.morf Exp.lex Exp.morf Exp.lex Exp.morf Exp.lex Exp.morf Exp.lex infelice-triste 4 4 0.48 0.52 0.525 0.687 0.125 0.137 impreciso-approssimativo 2 2 0.57 0.43 0.256 0.833 0.081 0.222 scorretto-sbagliato 3 3 0.48 0.52 0.025 0.986 0.076 0.144 imprudente-avventato 2 1 0.43 0.57 0.437 0.637 0.012 0.075 insufficiente-scarso 3 3 0.40 0.60 0.743 0.662 0.012 0.037 Table 2 Final experiment results. Number of senses, average of relative frequency and average of the accuracy of the two tasks: mask token prediction and polarity flip. towards either the lexical antonym or the morpholog- range of senses (see, e.g., Murphy [4]). ical antonym. Some nouns form with the adjective a We believe that these results, that contradict our ini- fairly stable collocation, while other nouns form freer ex- tial hypothesis, open up new avenues for future research pressions. For the purposes of this study, particularly in in this area, despite the limitations of the present study, relation to the analysis of competition, the central nouns which has an exploratory nature and a narrow empiri- with similar frequencies are of greater importance. cal coverage. Indeed, only 5 adjectives were analyzed, Upon examining the occurrences and accuracy, we exclusively belonging to core vocabulary. Another short- observe that the values are comparable. coming is that, unlike for English, there are no in-depth As for masked-token prediction, the consistent higher studies on antonyms in Italian. However, we want to values of the lexical antonym indicate higher predictabil- stress the importance of conducting studies on languages ity and/or higher degree of idiomaticity, which contra- other than English to avoid the well-known Anglocentric dicts our working hypothesis that lexical antonyms dis- bias. play a broader semantic coverage. Hopefully, our results will be challenged by further Finally, as for the polarity flip, consistently with the studies in the future, which might even overturn our con- masked-token experiments, the sentences containing the clusions entirely, if a larger data set were considered. Fur- morphological antonym were turned into positive using thermore, it would be interesting to investigate whether the lexical base fewer times than their lexical counter- the results obtained for Italian are also found for other parts, suggesting that the latter may have a more re- languages that present both lexical and morphological stricted semantic spectrum, contrary to our initial hy- antonyms, including languages with a different morpho- pothesis.7 logical system. With a view of deepening the analysis methodologically, it would be interesting to focus on additional linguistic factors that might drive the choice 7. Conclusion and Future Work between lexical and morphological antonyms, such as semantic networks or word frequency, and to expand the Our study investigated the differences and competition testing to the psycholinguistic dimension. between two types of antonyms, morphological and lexi- What is sure is that the relationship between mor- cal, focusing on a computational account of their context phological and lexical antonyms is more complex than of use. While a lexical analysis did not prove decisive, previously thought and that the choice of one type of experiments on masked-token prediction and polarity antonym over another depends on a variety of intercon- flip, aimed at approximating their semantic coverage, in- nected factors that are still to be fully unveiled. dicate that, unlike what is suggested in some studies on antonymy, the lexical antonym seems to have a narrower lexical coverage and scope, supporting the view that it References is actually the morphological antonym, despite its closer relationship with the lexical base, that displays a wider [1] S. Jones, M. L. Murphy, C. Paradis, C. Willners, Antonyms in English: Construals, Constructions 7 An anonymous reviewer observes that this result is even more re- and Canonicity, Cambridge University Press, 2012. markable given the potential purely morphological (rather than [2] D. A. Cruse, Lexical semantics, Cambridge Univer- semantic) bias due to the derivational relatedness of the morpho- sity Press, 1986. logical antonym, that could be predicted to favour its replacement by the target positive adjective to some degree. [3] E. Sapir, Grading, a study in semantics., Philosophy [20] C. Iacobini, Prefissazione, in: M. G. e F. Rainer (Ed.), of Science 11(2) (1944) 93–116. La formazione delle parole in italiano, Tübingen: [4] M. L. Murphy, Semantic relations and the lexicon, Niemeyer, 2004, pp. 97–163. Cambridge University Press, 2003. [21] G. Pittàno, Il grande dizionario dei Sinonimi e dei [5] C. Paradis, C. Willners, Antonymy and negation - Contrari, Zanichelli, 2013. the boundedness hypothesis, Journal of Pragmatics [22] N. Zingarelli, Zingarelli 2024. Vocabolario della lin- 38 (2006) 1051–1080. gua italiana, Zanichelli, 2023. [6] D. S. Palermo, J. J. Jenkins, Word association norms: [23] M. Aronoff, Competition and the lexicon, in: E. An- grade school through college, Minneapolis: Univer- nibale, C. Iacobini, M. Voghera (Eds.), Livelli di anal- sity of Minnesota Press, 1964. isi e fenomeni di interfaccia, Roma: Bulzoni, 2016. [7] J. Deese, The structure of associations in language [24] A. Niwa, K. Nishiguchi, N. Okazaki, Predicting and thought, Baltimore: Johns Hopkins University antonyms in context using BERT, in: A. Belz, Press, 1965. A. Fan, E. Reiter, Y. Sripada (Eds.), Proceedings [8] W. G. Charles, G. A. Miller, Contexts of antonymous of the 14th International Conference on Natural adjectives, Applied Psycholinguistics 10 (1989) Language Generation, Association for Computa- 357–375. tional Linguistics, Aberdeen, Scotland, UK, 2021, pp. [9] D. J. Hermann, G. Conti, D. Peters, P. H. Robbins, 48–54. URL: https://aclanthology.org/2021.inlg-1.6. R. J. S. Chaffin, Comprehension of antonymy and doi:10.18653/v1/2021.inlg- 1.6 . the generality of categorization models, Journal of Experimental Psychology: Human Learning and Memory 5 (1979) 585–597. [10] D. Gross, U. Fischer, G. A. Miller, The organization of adjectival meanings, Journal of Memory and Language 28 (1989) 92–106. [11] S. Jones, Antonymy: a corpus-based approach, Routledge, 2002. [12] J. S. Justeson, S. M. Katz, Co-occurrences of antony- mous adjectives and their contexts, Computational Linguistics 17 (1991) 1–20. URL: https://aclantholo gy.org/J91-1001. [13] L. Aina, R. Bernardi, R. Fernández, Negated adjec- tives and antonyms in distributional semantics: not similar, in: Proceedings of the Fifth Italian Confer- ence on Computational Linguistics, IJCoL, 2019, pp. 57–71. URL: http://journals.openedition.org/ijcol/4 57. doi:https://doi.org/10.4000/ijcol.457 . [14] M. Koptjevskaja-Tamm, M. Miestamo, C. Börstell, A cross-linguistic study of lexical and derived antonymy, Linguistics (2024). URL: https://doi. org/10.1515/ling-2023-0140. doi:doi:10.1515/li ng- 2023- 0140 . [15] F. Vicario, Note sull’ordine degli elementi in coppie di verbi antonimi., Linguistica XLIII 43 (2003) 3–12. [16] V. Muehleisen, Antonymy and Semantic range in English, Ph.D. thesis, Northwestern University, Evanston, 1997. [17] G. K. Zipf, Human behavior and the principle of least effort, Cambridge Addison-Wesley, 1949. [18] R. von Jhering, Der Zweck im Recht, Leipzig: Bre- itkopf and Hӓrtel, 1883. [19] K. E. Zimmer, Affixal negation in english and other languages: an investigation of restricted productiv- ity, Word. Journal of the Linguistic Circle of New York 20 (1964). APPENDIX A accuracy accuracy accuracy accuracy Noun infelice triste mtp (M) mtp (L) pf (M) pf (L) matrimonio 886 24 0.9 0.5 0.2 0.4 scelta 821 39 0.8 0.5 0 0.1 adolescenza 33 16 1 0.2 0 0.2 donna 325 316 0.5 0.8 0.2 0.1 uomo 300 440 0.5 0.6 0.4 0.1 situazione 120 339 0.4 0.9 0 0 momento 145 2.686 0.1 1 0 0.2 pagina 19 544 0 1 0.2 0 accuracy accuracy accuracy accuracy Noun impreciso approssimativo mtp (M) mtp (L) pf (M) pf (L) affermazione 64 11 0.2 0.8 0.1 0.1 notizia 223 58 0.4 0.8 0 0.1 terminologia 23 17 0.3 0.7 0.1 0.2 ricezione 50 18 0.7 0.3 0.2 0.1 misurazione 31 52 0.4 0.9 0.1 0.5 traduzione 54 235 0.2 0.7 0 0 conoscenza 33 226 0.1 0.9 0 0 calcolo 23 1.305 0.2 1 0.1 0.6 accuracy accuracy accuracy accuracy Noun scorretto sbagliato mtp (M) mtp (L) pf (M) pf (L) gioco 1.124 72 0.1 0.7 0 0 uso 2.901 276 0 1 0 0.3 alimentazione 2.926 1.228 0 0.9 0 0.1 posizione 1.317 910 0.1 1 0.3 0.2 abitudine 648 1.077 0.1 1 0.1 0 informazione 720 1.903 0 1 0.2 0.3 mossa 90 741 0 1 0 0.1 messaggio 88 1.214 0.1 0.9 0 0.1 accuracy accuracy accuracy accuracy Noun imprudente avventato mtp (M) mtp (L) pf (M) pf (L) condotta 394 14 0.2 0.8 0 0.2 comportamento 680 77 0.5 0.8 0 0 parola 49 46 0.3 0.5 0 0 manovra 24 35 0.5 0.6 0.1 0.4 gesto 54 195 0.7 0.4 0 0 azione 63 245 0.3 0.8 0 0 scelta 73 373 0.5 0.7 0 0 decisione 24 478 0.5 0.5 0 0 accuracy accuracy accuracy accuracy Noun insufficiente scarso mtp (M) mtp (L) pf (M) pf (L) apporto 263 19 0.9 0.4 0 0.2 quantità 1.097 162 1 0.4 0.1 0 alimentazione 300 120 1 0.5 0 0 produzione 208 158 0.9 0.4 0 0 utilizzo 16 27 0.9 0.9 0 0 pulizia 63 148 0.4 0.8 0 0 partecipazione 13 37 0.5 0.9 0 0 visibilità 14 274 0.4 1 0 0.1 Table 3 Co-occurrence frequencies of noun + morphological antonym and noun + lexical antonym. Accuracy of the two task: mtp (Masked-Token Prediction) and pf (Polarity Flip) related to Morphological Antonyms (M) and Lexical Antonyms (L). morphological antonym TTR lexical antonym TTR infelice 0.4694864048 triste 0.4504979496 impreciso 0.4726656991 approssimativo 0.4814814815 scorretto 0.4496086106 sbagliato 0.4500775996 imprudente 0.476119403 avventato 0.4644572526 insufficiente 0.4582118562 scarso 0.4805725971 Table 4 Token Type Ratio of 5 antonym pair from sentences extracted from itTenTen20 APPENDIX B system_message = '''In una frase l'aggettivo originale è stato sostituito da un token [MASK]. Tu devi riscrivere la frase facendo minimi cambiamenti e sostituire l'aggettivo mascherato con un altro aggettivo, in modo che la frase risulti volta al positivo. Il tuo output deve essere SOLO un json nel seguente formato e con i seguenti campi: {"new_sentence": "", "new_adj": ""}''' user_message = f'''Frase originale: "{masked_sent}" (aggettivo originale: {agg})''' {system_message = f'''In a sentence, the original adjective has been replaced by a [MASK] token. You need to rewrite the sentence making minimal changes and replace the masked adjective with another adjective, so that the sentence is positively oriented. Your output must be ONLY a json in the following format and with the following fields:\n''' + '''{"new\_sentence": "< your new sentence>", "new\_adj": "