Cover Image

Token embeddings violate the manifold hypothesis

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Robinson, Michael, Dey, Sourya, Chiang, Tony
Format:	Preprint
Published:	2025
Subjects:	Computation and Language Artificial Intelligence 53Z50, 62H15
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

The structure of the token space for large language models
by: Robinson, Michael, et al.
Published: (2024)

Probing the topology of the space of tokens with structured prompts
by: Robinson, Michael, et al.
Published: (2025)

Fast $k$-means clustering in Riemannian manifolds via Fréchet maps: Applications to large-dimensional SPD matrices
by: Shi, Ji, et al.
Published: (2025)

On warped product on information geometry: statistical manifolds and statistical models
by: Alba, Nicolás Martínez, et al.
Published: (2024)

Curvature-driven manifold fitting under unbounded isotropic noise
by: Li, Ruowei, et al.
Published: (2026)

On empirical Hodge Laplacians under the manifold hypothesis
by: Lerch, Jan-Paul, et al.
Published: (2025)

Why Agent Caching Fails and How to Fix It: Structured Intent Canonicalization with Few-Shot Learning
by: Basu, Abhinaba
Published: (2026)

TED++: Submanifold-Aware Backdoor Detection via Layerwise Tubular-Neighbourhood Screening
by: Le, Nam, et al.
Published: (2025)

Autoregressive Processes on Stiefel and Grassmann Manifolds
by: Figueras, Jordi-Lluís, et al.
Published: (2025)

Adaptive Stochastic Gradient Descents on Manifolds with an Application on Weighted Low-Rank Approximation
by: Yang, Peiqi, et al.
Published: (2025)

A Hamiltonian formalism for general variational problems, with applications to first order gravity with basis
by: Quijón, Guadalupe, et al.
Published: (2024)

Bi-invariant Geodesic Regression with Data from the Osteoarthritis Initiative
by: Schade, Johannes, et al.
Published: (2025)

A note on the theoretical approach to Grassmannians and Plücker coordinates for additive skew-symmetric pairwise comparisons matrices
by: Koczkodaj, Waldemar W., et al.
Published: (2025)

Bounds on the geodesic distances on the Stiefel manifold for a family of Riemannian metrics
by: Mataigne, Simon, et al.
Published: (2024)

Evaluating Embedding Generalization: How LLMs, LoRA, and SLERP Shape Representational Geometry
by: Kabane, Siyaxolisa
Published: (2025)

A Gauss-Bonnet-Chern type obstruction for Killing vector fields on Lorentzian manifolds
by: Romero, Alfonso, et al.
Published: (2024)

Convergence of Riemannian Stochastic Gradient Descents: Varying Batch Sizes And Nonstandard Batch Forming
by: Wu, Hao
Published: (2026)

Convergent autoencoder approximation of low bending and low distortion manifold embeddings
by: Braunsmann, Juliane, et al.
Published: (2022)

Optimization on the Extended Tensor-Train Manifold with Shared Factors
by: Molozhavenko, Alexander, et al.
Published: (2025)

Learning Dynamical Systems with the Spectral Exterior Calculus
by: Das, Suddhasattwa, et al.
Published: (2025)

Mapping the Web of Science, a large-scale graph and text-based dataset with LLM embeddings
by: Kunt, Tim, et al.
Published: (2026)

ConicCurv: A curvature estimation algorithm for planar polygons
by: Fuentes, R. Díaz, et al.
Published: (2025)

Bi-invariant Dissimilarity Measures for Sample Distributions in Lie Groups
by: Hanik, Martin, et al.
Published: (2024)

Characterization of geodesic completeness for landmark space
by: Habermann, Karen, et al.
Published: (2025)

An efficient algorithm for the Riemannian logarithm on the Stiefel manifold for a family of Riemannian metrics
by: Mataigne, Simon, et al.
Published: (2024)

Computing equivariant matrices on homogeneous spaces for Geometric Deep Learning and Automorphic Lie Algebras
by: Knibbeler, Vincent
Published: (2023)

ICE: Intervention-Consistent Explanation Evaluation with Statistical Grounding for LLMs
by: Basu, Abhinaba, et al.
Published: (2026)

MEDLEY-BENCH: Scale Buys Evaluation but Not Control in AI Metacognition
by: Abtahi, Farhad, et al.
Published: (2026)

An new polar factor retraction on the Stiefel manifold with closed-form inverse
by: Jensen, Rasmus, et al.
Published: (2026)

Boundary Detection Algorithm Inspired by Locally Linear Embedding
by: Kuo, Pei-Cheng, et al.
Published: (2024)

Evaluation of the number of clusters in a data set using $p$-values from Multiple Tests of Hypotheses
by: Modak, Soumita
Published: (2026)

Characterization of generalized quasi-Einstein manifolds and modified gravity
by: De, Uday Chand, et al.
Published: (2025)

A Survey of Dimension Estimation Methods
by: Binnie, James A. D., et al.
Published: (2025)

On the Gaussian Kinematic Formula of R. Adler and J. Taylor
by: Fu, Joseph H. G.
Published: (2024)

Proximity Measure of Information Object Features for Solving the Problem of Their Identification in Information Systems
by: Yuzefovych, Volodymyr
Published: (2026)

MindScape Study: Integrating LLM and Behavioral Sensing for Personalized AI-Driven Journaling Experiences
by: Nepal, Subigya, et al.
Published: (2024)

On the Equilibrium Locus of a Parameterized Dynamical System with Independent First Integrals
by: Kaminski, Yirmeyahu, et al.
Published: (2025)

Isometric simplices for high-order anisotropic mesh adaptation. Part I: Definition and existence of isometric triangulations
by: Bawin, Arthur, et al.
Published: (2025)

On the approximation of the Riemannian barycenter
by: Mataigne, Simon, et al.
Published: (2025)

A visual observation on the geometry of UMAP projections of the difference vectors of antonym and synonym word pair embeddings
by: Luisto, Rami
Published: (2026)