Geometric Patterns of Meaning: A PHATE Manifold Analysis of Multi-lingual Embeddings
We introduce a multi-level analysis framework for examining semantic geometry in multilingual embeddings, implemented through Semanscope (a visualization tool that applies PHATE manifold learning across four linguistic levels). Analysis of diverse datasets spanning sub-character components, alphabetic systems, semantic domains, and numerical concepts reveals systematic geometric patterns and critical limitations in current embedding models. At the sub-character level, purely structural elements (Chinese radicals) exhibit geometric collapse, highlighting model failures to distinguish semantic from structural components. At the character level, different writing systems show distinct geometric signatures. At the word level, content words form clustering-branching patterns across 20 semantic domains in English, Chinese, and German. Arabic numbers organize through spiral trajectories rather than clustering, violating standard distributional semantics assumptions. These findings establish PHATE manifold learning as an essential analytic tool not only for studying geometric structure of meaning in embedding space, but also for validating the effectiveness of embedding models in capturing semantic relationships.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Geometric Structures and Patterns of Meaning: A PHATE Manifold Analysis of Chinese Character Embeddings
We systematically investigate geometric patterns in Chinese character embeddings using PHATE manifold analysis. Through cross-validation across seven embedding models and eight dimensionality reduction methods, we observ…
Dimensionality ReductionExploring epithelial-cell calcium signaling with geometric and topological data analysis
Spatio-temporal calcium imaging is widely used to study neuronal activity. However the function of calcium signaling in epithelial cells is not well understood. We thus explore the underlying fluorescence patterns using …
Topological Data AnalysisKinetic Analysis of Protein Assembly on the Basis of the Center Manifold around the Critical Point
Protein assembly plays an important role in the regulation of biological systems. The cytoskeleton assembly activity is provided by the binding cofactors GTP (guanidine triphosphate) or ATP(adenosine triphosphate) to mon…
Exploring the Manifold of Neural Networks Using Diffusion Geometry
Drawing motivation from the manifold hypothesis, which posits that most high-dimensional data lies on or near low-dimensional manifolds, we apply manifold learning to the space of neural networks. We learn manifolds wher…
Dimensionality ReductionHyperparameter OptimizationNeural Architecture SearchPrincipal Curvatures Estimation with Applications to Single Cell Data
The rapidly growing field of single-cell transcriptomic sequencing (scRNAseq) presents challenges for data analysis due to its massive datasets. A common method in manifold learning consists in hypothesizing that dataset…