Imagen de Portada

Disentangling Direction and Magnitude in Transformer Representations: A Double Dissociation Through L2-Matched Perturbation Analysis

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Vardhan, Mangadoddi Srikar, Teja, Lekkala Sai
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language Artificial Intelligence Machine Learning I.2.7
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Weber's Law in Transformer Magnitude Representations: Efficient Coding, Representational Geometry, and Psychophysical Laws in Language Models
por: Cacioli, Jon-Paul
Publicado: (2026)

Cognitive Load Limits in Large Language Models: Benchmarking Multi-Hop Reasoning
por: Adapala, Sai Teja Reddy
Publicado: (2025)

Fine-Grained Detection of AI-Generated Text Using Sentence-Level Segmentation
por: Teja, Lekkala Sai, et al.
Publicado: (2025)

PSK at SemEval-2026 Task 9: Multilingual Polarization Detection Using Ensemble Gemma Models with Synthetic Data Augmentation
por: Pulipaka, Srikar Kashyap
Publicado: (2026)

BitCal-TTS: Bit-Calibrated Test-Time Scaling for Quantized Reasoning Models
por: Patarlapalli, Sai Babu, et al.
Publicado: (2026)

Robustness of Large Language Models to Perturbations in Text
por: Singh, Ayush, et al.
Publicado: (2024)

Give it Space! Explicit Disentangling of Positional and Semantic Representations in Encoders
por: Lequeu, Pierre-Antoine, et al.
Publicado: (2026)

The Anti-Ouroboros Effect: Emergent Resilience in Large Language Models from Recursive Selective Feedback
por: Adapala, Sai Teja Reddy
Publicado: (2025)

Listen to the Layers: Mitigating Hallucinations with Inter-Layer Disagreement
por: Subbalakshmi, Koduvayur, et al.
Publicado: (2026)

Bidirectional RAG: Safe Self-Improving Retrieval-Augmented Generation Through Multi-Stage Validation
por: Chinthala, Teja
Publicado: (2025)

Whether, Not Which: Mechanistic Interpretability Reveals Dissociable Affect Reception and Emotion Categorization in LLMs
por: Keeman, Michael
Publicado: (2026)

Seeing Through the Fog: A Cost-Effectiveness Analysis of Hallucination Detection Systems
por: Thomas, Alexander, et al.
Publicado: (2024)

RomanLens: The Role Of Latent Romanization In Multilinguality In LLMs
por: Saji, Alan, et al.
Publicado: (2025)

Text-Based Approaches to Item Difficulty Modeling in Large-Scale Assessments: A Systematic Review
por: Peters, Sydney, et al.
Publicado: (2025)

Enhancing Hate Speech Detection on Social Media: A Comparative Analysis of Machine Learning Models and Text Transformation Approaches
por: Mishra, Saurabh, et al.
Publicado: (2026)

The Dark Side of AI Transformers: Sentiment Polarization & the Loss of Business Neutrality by NLP Transformers
por: Kumar, Prasanna
Publicado: (2026)

Natural Language Processing for Tigrinya: Current State and Future Directions
por: Gaim, Fitsum, et al.
Publicado: (2025)

Exploring Graph Representations of Logical Forms for Language Modeling
por: Sullivan, Michael
Publicado: (2025)

Teaching Probabilistic Logical Reasoning to Transformers
por: Nafar, Aliakbar, et al.
Publicado: (2023)

Egalitarian Language Representation in Language Models: It All Begins with Tokenizers
por: Velayuthan, Menan, et al.
Publicado: (2024)

Direct Semantic Communication Between Large Language Models via Vector Translation
por: Yang, Fu-Chun, et al.
Publicado: (2025)

Mechanistic evaluation of Transformers and state space models
por: Arora, Aryaman, et al.
Publicado: (2025)

Exploring RWKV for Sentence Embeddings: Layer-wise Analysis and Baseline Comparison for Semantic Similarity
por: Pan, Xinghan
Publicado: (2025)

GNN-CNN: An Efficient Hybrid Model of Convolutional and Graph Neural Networks for Text Representation
por: Rastakhiz, Fardin
Publicado: (2025)

ChatGPT, Let us Chat Sign Language: Experiments, Architectural Elements, Challenges and Research Directions
por: Shahin, Nada, et al.
Publicado: (2024)

A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models
por: Rai, Daking, et al.
Publicado: (2024)

Assessment of Transformer-Based Encoder-Decoder Model for Human-Like Summarization
por: Nair, Sindhu, et al.
Publicado: (2024)

Transforming Dutch: Debiasing Dutch Coreference Resolution Systems for Non-binary Pronouns
por: van Boven, Goya, et al.
Publicado: (2024)

Low-Resource Court Judgment Summarization for Common Law Systems
por: Liu, Shuaiqi, et al.
Publicado: (2024)

Efficient Toxicity Detection in Gaming Chats: A Comparative Study of Embeddings, Fine-Tuned Transformers and LLMs
por: Tereshchenko, Yehor, et al.
Publicado: (2025)

Evaluating the Efficacy of Hybrid Deep Learning Models in Distinguishing AI-Generated Text
por: Oketunji, Abiodun Finbarrs
Publicado: (2023)

Weakly Supervised Distillation of Hallucination Signals into Transformer Representations
por: Salehmohamed, Shoaib Sadiq, et al.
Publicado: (2026)

StyloAI: Distinguishing AI-Generated Content with Stylometric Analysis
por: Opara, Chidimma
Publicado: (2024)

Dynamic Domain Information Modulation Algorithm for Multi-domain Sentiment Analysis
por: Yue, Chunyi, et al.
Publicado: (2025)

ELECTRA and GPT-4o: Cost-Effective Partners for Sentiment Analysis
por: Beno, James P.
Publicado: (2024)

MultiMatch: Multihead Consistency Regularization Matching for Semi-Supervised Text Classification
por: Sirbu, Iustin, et al.
Publicado: (2025)

How Instruction-Tuning Imparts Length Control: A Cross-Lingual Mechanistic Analysis
por: Rocchetti, Elisabetta, et al.
Publicado: (2025)

Internal Reasoning vs. External Control: A Thermodynamic Analysis of Sycophancy in Large Language Models
por: Chang, Edward Y.
Publicado: (2025)

A Lightweight Approach to Detection of AI-Generated Texts Using Stylometric Features
por: Aityan, Sergey K., et al.
Publicado: (2025)

SD$^2$: Self-Distilled Sparse Drafters
por: Lasby, Mike, et al.
Publicado: (2025)