Imagen de Portada

Tomato, Tomahto, Tomate: Do Multilingual Language Models Understand Based on Subword-Level Semantic Concepts?

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Zhang, Crystina, Lu, Jing, Tran, Vinh Q., Schuster, Tal, Metzler, Donald, Lin, Jimmy
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 Role of Mixed-Language Documents for Multilingual Large Language Model Pretraining
por: Shao, Jiandong, et al.
Publicado: (2026)

Hard Negatives, Hard Lessons: Revisiting Training Data Quality for Robust Information Retrieval with LLMs
por: Thakur, Nandan, et al.
Publicado: (2025)

Understanding Subword Compositionality of Large Language Models
por: Peng, Qiwei, et al.
Publicado: (2025)

StochasTok: Improving Fine-Grained Subword Understanding in LLMs
por: Sims, Anya, et al.
Publicado: (2025)

A Systematic Analysis of Subwords and Cross-Lingual Transfer in Multilingual Translation
por: Meyer, Francois, et al.
Publicado: (2024)

CSF: Contrastive Semantic Features for Direct Multilingual Sign Language Generation
por: Bao, Tran Sy
Publicado: (2026)

LACONIC: Dense-Level Effectiveness for Scalable Sparse Retrieval via a Two-Phase Training Curriculum
por: Xu, Zhichao, et al.
Publicado: (2026)

OFA: A Framework of Initializing Unseen Subword Embeddings for Efficient Large-scale Multilingual Continued Pretraining
por: Liu, Yihong, et al.
Publicado: (2023)

Impact of Preference Noise on the Alignment Performance of Generative Language Models
por: Gao, Yang, et al.
Publicado: (2024)

Rankers, Judges, and Assistants: Towards Understanding the Interplay of LLMs in Information Retrieval Evaluation
por: Balog, Krisztian, et al.
Publicado: (2025)

Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?
por: Li, Xinzhe, et al.
Publicado: (2023)

Drawing Conclusions from Draws: Rethinking Preference Semantics in Arena-Style LLM Evaluation
por: Tang, Raphael, et al.
Publicado: (2025)

SEMQA: Semi-Extractive Multi-Source Question Answering
por: Schuster, Tal, et al.
Publicado: (2023)

Do Language Models' Words Refer?
por: Mandelkern, Matthew, et al.
Publicado: (2023)

Fractal Patterns May Illuminate the Success of Next-Token Prediction
por: Alabdulmohsin, Ibrahim, et al.
Publicado: (2024)

The Learning Dynamics of Subword Segmentation for Morphologically Diverse Languages
por: Meyer, Francois, et al.
Publicado: (2025)

Morphological Typology in BPE Subword Productivity and Language Modeling
por: Parra, Iñigo
Publicado: (2024)

Evaluating Subword Tokenization: Alien Subword Composition and OOV Generalization Challenge
por: Batsuren, Khuyagbaatar, et al.
Publicado: (2024)

Improving Multilingual Math Reasoning for African Languages
por: Ogundepo, Odunayo, et al.
Publicado: (2025)

Distributional Properties of Subword Regularization
por: Cognetta, Marco, et al.
Publicado: (2024)

Lexically Grounded Subword Segmentation
por: Libovický, Jindřich, et al.
Publicado: (2024)

Tokenization Falling Short: On Subword Robustness in Large Language Models
por: Chai, Yekun, et al.
Publicado: (2024)

All Languages Matter: Understanding and Mitigating Language Bias in Multilingual RAG
por: Wang, Dan, et al.
Publicado: (2026)

On the Effect of (Near) Duplicate Subwords in Language Modelling
por: Schäfer, Anton, et al.
Publicado: (2024)

Entailment Semantics Can Be Extracted from an Ideal Language Model
por: Merrill, William, et al.
Publicado: (2022)

ByteSpan: Information-Driven Subword Tokenisation
por: Goriely, Zébulon, et al.
Publicado: (2025)

Subword Tokenization Strategies for Kurdish Word Embeddings
por: Salehi, Ali, et al.
Publicado: (2025)

Transferring Extreme Subword Style Using Ngram Model-Based Logit Scaling
por: Messner, Craig, et al.
Publicado: (2025)

MINERS: Multilingual Language Models as Semantic Retrievers
por: Winata, Genta Indra, et al.
Publicado: (2024)

Decoupling the Benefits of Subword Tokenization for Language Model Training via Byte-level Simulation
por: Gigant, Théo, et al.
Publicado: (2026)

How Do Multilingual Language Models Remember Facts?
por: Fierro, Constanza, et al.
Publicado: (2024)

Do Pre-Trained Language Models Detect and Understand Semantic Underspecification? Ask the DUST!
por: Wildenburg, Frank, et al.
Publicado: (2024)

LGSE: Lexically Grounded Subword Embedding Initialization for Low-Resource Language Adaptation
por: Teklehaymanot, Hailay, et al.
Publicado: (2026)

Subword models struggle with word learning, but surprisal hides it
por: Bunzeck, Bastian, et al.
Publicado: (2025)

Learning Mutually Informed Representations for Characters and Subwords
por: Wang, Yilin, et al.
Publicado: (2023)

Leading Whitespaces of Language Models' Subword Vocabulary Pose a Confound for Calculating Word Probabilities
por: Oh, Byung-Doh, et al.
Publicado: (2024)

Smaller, Weaker, Yet Better: Training LLM Reasoners via Compute-Optimal Sampling
por: Bansal, Hritik, et al.
Publicado: (2024)

Existential Definability over the Subword Ordering
por: Baumann, Pascal, et al.
Publicado: (2022)

INCLUDE: Evaluating Multilingual Language Understanding with Regional Knowledge
por: Romanou, Angelika, et al.
Publicado: (2024)

Do Multilingual Large Language Models Mitigate Stereotype Bias?
por: Nie, Shangrui, et al.
Publicado: (2024)