Conceptual Networks for Cross-Linguistic Idiomatic Expressions:A Feature-Based Graph Approach
2026-07-10 • Computation and Language
Computation and LanguageArtificial IntelligenceEmerging Technologies
AI summaryⓘ
The authors created a new way to represent the meanings of idioms and figurative expressions from eight different languages using networks based on shared conceptual features like emotions or containment. They found that idioms group together by these underlying ideas rather than by language, matching predictions from cognitive linguistics. Their method works better than some existing models for detecting idioms and helps find good translations between languages. They also showed that different types of features all add useful information and that the network structure itself helps understanding idioms. This approach provides a clear and generalizable way to study idiomatic meaning across languages.
IdiomsCognitive LinguisticsConceptual FeaturesNetwork AnalysisJaccard SimilarityCommunity DetectionCross-lingual TransferDistributional EmbeddingsLarge Language ModelsValence
Authors
Kiran Pala, Punam Silu, Lixun Yu
Abstract
We present an interpretable network-based framework for representing idiomatic and figurative meaning across eight typologically diverse languages, totaling 160 conventional expressions, the large majority of which are idiomatic. Each expression is annotated with binary conceptual features (containment, concealment, emotional, social, etc.) derived from cognitive-linguistic theory, and pairwise Jaccard similarities define a weighted graph. Community detection reveals that idioms cluster by conceptual schema rather than by language, producing a structure consistent with cognitive-linguistic predictions. The conceptual network captures unique semantic information not present in distributional embeddings, can be scaled via automatic annotation with LLMs, improves downstream idiom detection, and remains robust when enriched with corpus frequencies. Cross-lingual transfer experiments show that conceptual proximity alone can identify acceptable translation equivalents across five language families, with substantial gains over embedding-based baselines. Ablation studies demonstrate that all three feature dimensions -- schemas, roles, and valence -- contribute non-redundantly to both the network's organizational properties and its performance on idiom detection, and that specific graph-derived signals (community membership, neighbor similarity) are particularly informative. The framework offers an interpretable, cross-linguistically stable representation of idiomatic meaning, combining theoretical grounding with practical utility.