Imagen de Portada

Interpreting token compositionality in LLMs: A robustness analysis

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Aljaafari, Nura, Carvalho, Danilo S., Freitas, André
Formato:	Preprint
Publicado:	2024
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

The Mechanics of Conceptual Interpretation in GPT Models: Interpretative Insights
por: Aljaafari, Nura, et al.
Publicado: (2024)

TRACE: Training and Inference-Time Interpretability Analysis for Language Models
por: Aljaafari, Nura, et al.
Publicado: (2025)

Emergence and Localisation of Semantic Role Circuits in LLMs
por: Aljaafari, Nura, et al.
Publicado: (2025)

CARMA: Enhanced Compositionality in LLMs via Advanced Regularisation and Mutual Information Alignment
por: Aljaafari, Nura, et al.
Publicado: (2025)

TRACE for Tracking the Emergence of Semantic Representations in Transformers
por: Aljaafari, Nura, et al.
Publicado: (2025)

Is Inference Mediated by Distinct Semantic Structures in LLMs? A Mechanistic Interpretation
por: Aljaafari, Nura, et al.
Publicado: (2026)

From Circuit Evidence to Mechanistic Theory: An Inductive Logic Approach
por: Aljaafari, Nura, et al.
Publicado: (2026)

Bridging Compositional and Distributional Semantics: A Survey on Latent Semantic Geometry via AutoEncoder
por: Zhang, Yingji, et al.
Publicado: (2025)

Inductive Learning of Logical Theories with LLMs: An Expressivity-Graded Analysis
por: Gandarela, João Pedro, et al.
Publicado: (2024)

Quasi-symbolic Semantic Geometry over Transformer-based Variational AutoEncoder
por: Zhang, Yingji, et al.
Publicado: (2022)

Learning Disentangled Semantic Spaces of Explanations via Invertible Neural Networks
por: Zhang, Yingji, et al.
Publicado: (2023)

Multi-Relational Hyperbolic Word Embeddings from Natural Language Definitions
por: Valentino, Marco, et al.
Publicado: (2023)

Learning to Disentangle Latent Reasoning Rules with Language VAEs: A Systematic Study
por: Zhang, Yingji, et al.
Publicado: (2025)

Why do LLMs attend to the first token?
por: Barbero, Federico, et al.
Publicado: (2025)

Towards Controllable Natural Language Inference through Lexical Inference Types
por: Zhang, Yingji, et al.
Publicado: (2023)

LangVAE and LangSpace: Building and Probing for Language Model VAEs
por: Carvalho, Danilo S., et al.
Publicado: (2025)

SylloBio-NLI: Evaluating Large Language Models on Biomedical Syllogistic Reasoning
por: Wysocka, Magdalena, et al.
Publicado: (2024)

Montague semantics and modifier consistency measurement in neural language models
por: Carvalho, Danilo S., et al.
Publicado: (2022)

AnomaLLMy -- Detecting anomalous tokens in black-box LLMs through low-confidence single-token predictions
por: Witold, Waligóra
Publicado: (2024)

Tokenization counts: the impact of tokenization on arithmetic in frontier LLMs
por: Singh, Aaditya K., et al.
Publicado: (2024)

Prediction hubs are context-informed frequent tokens in LLMs
por: Nielsen, Beatrix M. G., et al.
Publicado: (2025)

Jacobian Scopes: token-level causal attributions in LLMs
por: Liu, Toni J. B., et al.
Publicado: (2026)

Finetuning LLMs for EvaCun 2025 token prediction shared task
por: Jon, Josef, et al.
Publicado: (2025)

Comparative analysis of subword tokenization approaches for Indian languages
por: Das, Sudhansu Bala, et al.
Publicado: (2025)

Improving Semantic Control in Discrete Latent Spaces with Transformer Quantized Variational Autoencoders
por: Zhang, Yingji, et al.
Publicado: (2024)

PEIRCE: Unifying Material and Formal Reasoning via LLM-Driven Neuro-Symbolic Refinement
por: Quan, Xin, et al.
Publicado: (2025)

Interpretable Next-token Prediction via the Generalized Induction Head
por: Kim, Eunji, et al.
Publicado: (2024)

Is my model "mind blurting"? Interpreting the dynamics of reasoning tokens with Recurrence Quantification Analysis (RQA)
por: Pham, Quoc Tuan, et al.
Publicado: (2026)

Accelerating Antibiotic Discovery with Large Language Models and Knowledge Graphs
por: Delmas, Maxime, et al.
Publicado: (2025)

Autoformalization in the Wild: Assessing LLMs on Real-World Mathematical Definitions
por: Zhang, Lan, et al.
Publicado: (2025)

Towards Nepali-language LLMs: Efficient GPT training with a Nepali BPE tokenizer
por: Shrestha, Adarsha, et al.
Publicado: (2025)

Where is the signal in tokenization space?
por: Geh, Renato Lui, et al.
Publicado: (2024)

Reasoning Circuits in Language Models: A Mechanistic Interpretation of Syllogistic Inference
por: Kim, Geonhee, et al.
Publicado: (2024)

Contextual morphologically-guided tokenization for Latin encoder models
por: Hudspeth, Marisa, et al.
Publicado: (2025)

You only need 4 extra tokens: Synergistic Test-time Adaptation for LLMs
por: Xu, Yijie, et al.
Publicado: (2025)

DeepMLF: Multimodal language model with learnable tokens for deep fusion in sentiment analysis
por: Georgiou, Efthymios, et al.
Publicado: (2025)

The pitfalls of next-token prediction
por: Bachmann, Gregor, et al.
Publicado: (2024)

Looking beyond the next token
por: Thankaraj, Abitha, et al.
Publicado: (2025)

On multi-token prediction for efficient LLM inference
por: Mehra, Somesh, et al.
Publicado: (2025)

A Survey in Mathematical Language Processing
por: Meadows, Jordan, et al.
Publicado: (2022)