CAMMAR: Culture-Aware Matryoshka for Metaphorical Arabic Representations
Computation and LanguageArtificial Intelligence
Summary
The authors study how metaphors in Arabic carry cultural meaning and how existing language models mix up different types of information, calling this 'semantic smearing.' They created CAMMAR, a new way to organize meanings into layers for words, culture, and metaphors, inspired by a classical Arabic theory. Their method can measure how metaphorical a word use is by looking at the distances between these layers and works well when guided by examples. They also found that including Arabic word roots helps improve the system slightly, and they plan to release their data and tools for others to use.
Authors
Suzan Awinat, Alfonso Ortega del Puente
Abstract
Metaphor in Arabic is a culturally grounded mechanism for constructing meaning, encoding cultural knowledge that shapes interpretation. Yet current Arabic language models typically collapse lexical, cultural, and metaphorical information into a single representational space, a phenomenon we term "semantic smearing". We introduce CAMMAR (Culture-Aware Matryoshka for Metaphorical Arabic Representations), a representation learning framework that organizes meaning into nested lexical, cultural, and metaphorical embedding subspaces through a staged semantic curriculum. The design implements compositional principles of Al-Jurjani's theory of nazum, modeling figurative meaning as compositionally grounded in prior semantic relations, and yields a training-free geometric measure of metaphoricity based on the distance between lexical and metaphorical representations. Evaluated on a new span-annotated Arabic metaphor set as word-matched figurative/literal pairs, the geometric readout detects metaphor well above chance when the inter-layer geometry is shaped by paired supervision (AUC up to 0.84; figurative outscores its literal counterpart for the same word in 82.6\% of pairs), but sits at chance under an unsupervised domain contrast alone, a clean separation between a legible-under-supervision regime and a non-emergent one. A controlled ablation shows that grounding the lexical layer in morphological roots gives a small but consistent gain, an effect absent from direct probing that reflects the layer's quality as a measurement anchor. We will release the datasets, cultural concept inventory, and code upon acceptance.