2020/10/09 by Jun Yen Leung, Leung, Jun Yen, Guy Emerson +3
Computer Science · Social Sciences · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Language and cultural evolution #Natural Language Processing Techniques #Topic Modeling
paper · pdf · doi:10.48550/arxiv.2010.04755
openalex publication_date 2020/10/09 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Across languages, multiple consecutive adjectives modifying a noun (e.g. "the\nbig red dog") follow certain unmarked ordering rules. While explanatory\naccounts have been put forward, much of the work done in this area has relied\nprimarily on the intuitive judgment of native speakers, rather than on corpus\ndata. We present the first purely corpus-driven model of multi-lingual\nadjective ordering in the form of a latent-variable model that can accurately\norder adjectives across 24 different languages, even when the training and\ntesting languages are different. We utilize this novel statistical model to\nprovide strong converging evidence for the existence of universal,\ncross-linguistic, hierarchical adjective ordering tendencies.\n