In this article, we will elucidate how nominalizers and nominal linkers as homophonous morphemes have been developed from genitive case-markers via grammaticalization, with a special attention to the diachronic process of the development of no 'of' in Japanese and its comparison with the corresponding morphemes de 'of' in Chinese and of in English. We will identify multiple uses of these homophonous grammatical morphemes, including genitive case-markers, nominalizers, and two subtypes of (syntactic) linkers. Drawing on historical corpus data, we will show that different uses of Japanese no emerged in different historical periods, especially the two subtypes of linker. We will also demonstrate that Chinese and English lack the linker (type-1) and have a limited use of the linker (type-2). We argue that the diachronic developmental process of linkers in Japanese and their limited usages in Chinese and English reflect different ranges of secondary grammaticalization via horizontal reanalysis of functional heads. We will also argue that all the linker constructions involve subject-predicate inversion in the sense of den Dikken (2006) and/or the movement of a complement to a specifier in the sense of Kayne (1994). Different behaviors of overt linkers in different languages are argued to follow from microparameters on the category selection by the linker heads.

The attempt to utilize machine learning in procedural content generation (PCG) has been made in the past. In this survey paper, we investigate how generative artificial intelligence (AI), which saw a significant increase in interest in the mid-2010s, is being used for PCG. We review applications of generative AI for the creation of various types of content, including terrains, items, and even storylines. While generative AI is effective for PCG, building high-performance models requires not only handling customized content and ensuring quality and diversity, but also securing sufficient training data. For PCG research to advance further, addressing these challenges is essential. Thus, we also give special consideration to research that explores innovative generation techniques, model architectures, and approaches suited for limited-data scenarios.