Image de couverture de livre

The Geometry of Tokens in Internal Representations of Large Language Models

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Viswanathan, Karthik, Gardinazzi, Yuri, Panerai, Giada, Cazzaniga, Alberto, Biagetti, Matteo
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Persistent Topological Features in Large Language Models
par: Gardinazzi, Yuri, et autres
Publié: (2024)

Probing Geometry of Next Token Prediction Using Cumulant Expansion of the Softmax Entropy
par: Viswanathan, Karthik, et autres
Publié: (2025)

Sparse-Autoencoder-Guided Internal Representation Unlearning for Large Language Models
par: Yamashita, Tomoya, et autres
Publié: (2025)

The Linear Representation Hypothesis and the Geometry of Large Language Models
par: Park, Kiho, et autres
Publié: (2023)

Truth as a Trajectory: What Internal Representations Reveal About Large Language Model Reasoning
par: Damirchi, Hamed, et autres
Publié: (2026)

The representation landscape of few-shot learning and fine-tuning in large language models
par: Doimo, Diego, et autres
Publié: (2024)

The Geometry of Refusal in Large Language Models: Concept Cones and Representational Independence
par: Wollschläger, Tom, et autres
Publié: (2025)

Scale Determines Whether Language Models Organize Representation Geometry for Prediction
par: Xu, Weilun
Publié: (2026)

Attention Sinks as Internal Signals for Hallucination Detection in Large Language Models
par: Binkowski, Jakub, et autres
Publié: (2026)

From Token to Token Pair: Efficient Prompt Compression for Large Language Models in Clinical Prediction
par: Zhu, Mingcheng, et autres
Publié: (2026)

Temporal Tokenization Strategies for Event Sequence Modeling with Large Language Models
par: Liu, Zefang, et autres
Publié: (2025)

Superscopes: Amplifying Internal Feature Representations for Language Model Interpretation
par: Jacobi, Jonathan, et autres
Publié: (2025)

Efficient Temporal Tokenization for Mobility Prediction with Large Language Models
par: He, Haoyu, et autres
Publié: (2025)

Towards Linguistically-Aware and Language-Independent Tokenization for Large Language Models (LLMs)
par: Rahman, Abrar, et autres
Publié: (2024)

On the Origins of Linear Representations in Large Language Models
par: Jiang, Yibo, et autres
Publié: (2024)

Implicit Geometry of Next-token Prediction: From Language Sparsity Patterns to Model Representations
par: Zhao, Yize, et autres
Publié: (2024)

Counterfactual Token Generation in Large Language Models
par: Chatzi, Ivi, et autres
Publié: (2024)

Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models
par: Jain, Neel, et autres
Publié: (2024)

Turning Trash into Treasure: Accelerating Inference of Large Language Models with Token Recycling
par: Luo, Xianzhen, et autres
Publié: (2024)

PrefixQuant: Eliminating Outliers by Prefixed Tokens for Large Language Models Quantization
par: Chen, Mengzhao, et autres
Publié: (2024)

Tokens for Learning, Tokens for Unlearning: Mitigating Membership Inference Attacks in Large Language Models via Dual-Purpose Training
par: Tran, Toan, et autres
Publié: (2025)

Confidence Geometry Reveals Trace-Level Correctness in Large Language Model Reasoning
par: Liu, Shuo, et autres
Publié: (2026)

Token-Level Adversarial Prompt Detection Based on Perplexity Measures and Contextual Information
par: Hu, Zhengmian, et autres
Publié: (2023)

Token-level Accept or Reject: A Micro Alignment Approach for Large Language Models
par: Zhang, Yang, et autres
Publié: (2025)

Lossless Compression of Large Language Model-Generated Text via Next-Token Prediction
par: Mao, Yu, et autres
Publié: (2025)

Probabilistic Token Alignment for Large Language Model Fusion
par: Zeng, Runjia, et autres
Publié: (2025)

Compositional Steering of Large Language Models with Steering Tokens
par: Radevski, Gorjan, et autres
Publié: (2026)

Token-Efficient Leverage Learning in Large Language Models
par: Zeng, Yuanhao, et autres
Publié: (2024)

Language Modeling with Learned Meta-Tokens
par: Shah, Alok N., et autres
Publié: (2025)

Parallel Token Prediction for Language Models
par: Draxler, Felix, et autres
Publié: (2025)

InnerThoughts: Disentangling Representations and Predictions in Large Language Models
par: Chételat, Didier, et autres
Publié: (2025)

Taxonomy, Opportunities, and Challenges of Representation Engineering for Large Language Models
par: Wehner, Jan, et autres
Publié: (2025)

Can Large Language Models Generalize Procedures Across Representations?
par: Lin, Fangru, et autres
Publié: (2026)

Tracing the Representation Geometry of Language Models from Pretraining to Post-training
par: Li, Melody Zixuan, et autres
Publié: (2025)

Geometry-Aware Decoding with Wasserstein-Regularized Truncation and Mass Penalties for Large Language Models
par: Davoodi, Arash Gholami, et autres
Publié: (2026)

UniMoT: Unified Molecule-Text Language Model with Discrete Token Representation
par: Guo, Shuhan, et autres
Publié: (2024)

The Geometry of Categorical and Hierarchical Concepts in Large Language Models
par: Park, Kiho, et autres
Publié: (2024)

Revisiting Graph-Tokenizing Large Language Models: A Systematic Evaluation of Graph Token Understanding
par: Zhang, Zhongjian, et autres
Publié: (2026)

In-Context Symbolic Regression: Leveraging Large Language Models for Function Discovery
par: Merler, Matteo, et autres
Publié: (2024)

Large Language Models Encode Semantics and Alignment in Linearly Separable Representations
par: Saglam, Baturay, et autres
Publié: (2025)