Imagen de Portada

Discrete Semantic States and Hamiltonian Dynamics in LLM Embedding Spaces

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autor principal:	Laine, Timo Aukusti
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

The Quantum LLM: Modeling Semantic Spaces with Quantum Principles
por: Laine, Timo Aukusti
Publicado: (2025)

Semantic Wave Functions: Exploring Meaning in Large Language Models Through Quantum Formalism
por: Laine, Timo Aukusti
Publicado: (2025)

Quantum LLMs Using Quantum Computing to Analyze and Process Semantic Information
por: Laine, Timo Aukusti
Publicado: (2025)

Dynamic Neighborhood Construction for Structured Large Discrete Action Spaces
por: Akkerman, Fabian, et al.
Publicado: (2023)

ss-Mamba: Semantic-Spline Selective State-Space Model
por: Ye, Zuochen
Publicado: (2025)

Dynamic Large Concept Models: Latent Reasoning in an Adaptive Semantic Space
por: Qu, Xingwei, et al.
Publicado: (2025)

From Exact Hits to Close Enough: Semantic Caching for LLM Embeddings
por: Biton, Dvir David, et al.
Publicado: (2026)

Embedding Knowledge Graph in Function Spaces
por: Teyou, Louis Mozart Kamdem, et al.
Publicado: (2024)

ClusterKV: Manipulating LLM KV Cache in Semantic Space for Recallable Compression
por: Liu, Guangda, et al.
Publicado: (2024)

One Swallow Does Not Make a Summer: Understanding Semantic Structures in Embedding Spaces
por: Sun, Yandong, et al.
Publicado: (2025)

Scale-Consistent State-Space Dynamics via Fractal of Stationary Transformations
por: Yu, Geunhyeok, et al.
Publicado: (2026)

Causal Abstraction Learning based on the Semantic Embedding Principle
por: D'Acunto, Gabriele, et al.
Publicado: (2025)

Educational Cone Model in Embedding Vector Spaces
por: Ehara, Yo
Publicado: (2025)

Compelling ReLU Networks to Exhibit Exponentially Many Linear Regions at Initialization and During Training
por: Milkert, Max, et al.
Publicado: (2023)

SAINT: Attention-Based Policies for Discrete Combinatorial Action Spaces
por: Landers, Matthew, et al.
Publicado: (2025)

Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluation
por: Wang, Yiming, et al.
Publicado: (2024)

GeoDynamics: A Geometric State-Space Neural Network for Understanding Brain Dynamics on Riemannian Manifolds
por: Dan, Tingting, et al.
Publicado: (2026)

LLM Embeddings for Deep Learning on Tabular Data
por: Koloski, Boshko, et al.
Publicado: (2025)

On the Dynamics of Observation and Semantics
por: Li, Xiu
Publicado: (2026)

SemanticOpt: Towards LLM-Based Semantic Black-Box Optimization
por: Meindl, Jamison, et al.
Publicado: (2025)

Explaining, Verifying, and Aligning Semantic Hierarchies in Vision-Language Model Embeddings
por: Schwalbe, Gesina, et al.
Publicado: (2026)

Low-distortion and GPU-compatible Tree Embeddings in Hyperbolic Space
por: van Spengler, Max, et al.
Publicado: (2025)

Probing the Embedding Space of Transformers via Minimal Token Perturbations
por: Conti, Eddie, et al.
Publicado: (2025)

Non-Stationarity in the Embedding Space of Time Series Foundation Models
por: Choi, Jinmyeong, et al.
Publicado: (2026)

Discretizing Continuous Action Space with Unimodal Probability Distributions for On-Policy Reinforcement Learning
por: Zhu, Yuanyang, et al.
Publicado: (2024)

Breaking the Grid: Distance-Guided Reinforcement Learning in Large Discrete Action Spaces
por: Hoppe, Heiko, et al.
Publicado: (2026)

Decoupled Hierarchical Reinforcement Learning with State Abstraction for Discrete Grids
por: Xiao, Qingyu, et al.
Publicado: (2025)

Discrete State Diffusion Models: A Sample Complexity Perspective
por: Srikanth, Aadithya, et al.
Publicado: (2025)

Analyzing and Improving the Training Dynamics of Diffusion Models
por: Karras, Tero, et al.
Publicado: (2023)

Exploiting Temporal State Space Sharing for Video Semantic Segmentation
por: Hesham, Syed Ariff Syed, et al.
Publicado: (2025)

Intrinsic-Energy Joint Embedding Predictive Architectures Induce Quasimetric Spaces
por: Kobanda, Anthony, et al.
Publicado: (2026)

A Geometry-Aware Algorithm to Learn Hierarchical Embeddings in Hyperbolic Space
por: Wang, Zhangyu, et al.
Publicado: (2024)

ES-Merging: Biological MLLM Merging via Embedding Space Signals
por: Lee, Wonbin, et al.
Publicado: (2026)

Minimizing Hyperbolic Embedding Distortion with LLM-Guided Hierarchy Restructuring
por: Ayoughi, Melika, et al.
Publicado: (2025)

Artificial Neural Nets and the Representation of Human Concepts
por: Freiesleben, Timo
Publicado: (2023)

Q-function Decomposition with Intervention Semantics with Factored Action Spaces
por: Lee, Junkyu, et al.
Publicado: (2025)

Beyond Semantic Manipulation: Token-Space Attacks on Reward Models
por: Zhang, Yuheng, et al.
Publicado: (2026)

Learning Reconstructive Embeddings in Reproducing Kernel Hilbert Spaces via the Representer Theorem
por: Feito-Casares, Enrique, et al.
Publicado: (2026)

Temporal Dynamic Embedding for Irregularly Sampled Time Series
por: Kim, Mincheol, et al.
Publicado: (2025)

Stability and Discretization Error of State Space Model Neural Operators
por: Bendahi, Abderrahim, et al.
Publicado: (2026)