Copertina

Training Language Models with homotokens Leads to Delayed Overfitting

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Cosma, Adrian, Ruseti, Stefan, Radoi, Emilian, Dascalu, Mihai
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

The Strawberry Problem: Emergence of Character-level Understanding in Tokenized Language Models
di: Cosma, Adrian, et al.
Pubblicazione: (2025)

How Hard is this Test Set? NLI Characterization by Exploiting Training Dynamics
di: Cosma, Adrian, et al.
Pubblicazione: (2024)

Value-Aware Numerical Representations for Transformer Language Models
di: Dutulescu, Andreea, et al.
Pubblicazione: (2026)

What Makes a Good Doctor Response? A Study on Text-Based Telemedicine
di: Cosma, Adrian, et al.
Pubblicazione: (2026)

A Retrieval-Based Approach to Medical Procedure Matching in Romanian
di: Niculae, Andrei, et al.
Pubblicazione: (2025)

Neural Grammatical Error Correction for Romanian
di: Cotet, Teodor-Mihai, et al.
Pubblicazione: (2026)

RoMath: A Mathematical Reasoning Benchmark in Romanian
di: Cosma, Adrian, et al.
Pubblicazione: (2024)

Dr.Copilot: A Multi-Agent Prompt Optimized Assistant for Improving Patient-Doctor Communication in Romanian
di: Niculae, Andrei, et al.
Pubblicazione: (2025)

On Model and Data Scaling for Skeleton-based Self-Supervised Gait Recognition
di: Cosma, Adrian, et al.
Pubblicazione: (2025)

The Paradox of Motion: Evidence for Spurious Correlations in Skeleton-based Gait Recognition Models
di: Cătrună, Andy, et al.
Pubblicazione: (2024)

Spatial Colour Mixing Illusions as a Perception Stress Test for Vision-Language Models
di: Basoc, Nicoleta-Nina, et al.
Pubblicazione: (2026)

MoME: Estimating Psychological Traits from Gait with Multi-Stage Mixture of Movement Experts
di: Cǎtrunǎ, Andy, et al.
Pubblicazione: (2025)

CrossGaze: A Strong Method for 3D Gaze Estimation in the Wild
di: Cătrună, Andy, et al.
Pubblicazione: (2024)

GaitPT: Skeletons Are All You Need For Gait Recognition
di: Catruna, Andy, et al.
Pubblicazione: (2023)

Database-Agnostic Gait Enrollment using SetTransformers
di: Basoc, Nicoleta, et al.
Pubblicazione: (2025)

"Înţelegi Româneşte?'' A Recipe for Romanian Vision-Language Models
di: Masala, Mihai, et al.
Pubblicazione: (2026)

An In-Vitro Study on Cross-Lingual Generalization in Language Models
di: Cosma, Adrian
Pubblicazione: (2026)

Aligning Actions and Walking to LLM-Generated Textual Descriptions
di: Chivereanu, Radu, et al.
Pubblicazione: (2024)

Gait Recognition from Highly Compressed Videos
di: Niculae, Andrei, et al.
Pubblicazione: (2024)

Dual-objective Language Models: Training Efficiency Without Overfitting
di: Samuel, David, et al.
Pubblicazione: (2025)

Uncovering Overfitting in Large Language Model Editing
di: Zhang, Mengqi, et al.
Pubblicazione: (2024)

RoCode: A Dataset for Measuring Code Intelligence from Problem Definitions in Romanian
di: Cosma, Adrian, et al.
Pubblicazione: (2024)

"Vorbeşti Româneşte?" A Recipe to Train Powerful Romanian LLMs with English Instructions
di: Masala, Mihai, et al.
Pubblicazione: (2024)

Gumbel Machine: Counterfactual Student Writing Generation via Gumbel Noise Steering
di: McNichols, Hunter, et al.
Pubblicazione: (2026)

OpenLLM-Ro -- Technical Report on Open-source Romanian LLMs
di: Masala, Mihai, et al.
Pubblicazione: (2024)

Dialectal and Low-Resource Machine Translation for Aromanian
di: Jerpelea, Alexandru-Iulius, et al.
Pubblicazione: (2024)

Post-Training Language Models for Continual Relation Extraction
di: Efeoglu, Sefika, et al.
Pubblicazione: (2025)

Automatic Prompt Optimization for Dataset-Level Feature Discovery
di: Cosma, Adrian, et al.
Pubblicazione: (2026)

TF3-RO-50M: Training Compact Romanian Language Models from Scratch on Synthetic Moral Microfiction
di: Nadas, Mihai Dan, et al.
Pubblicazione: (2026)

Evaluating Large Language Models for Diacritic Restoration in Romanian Texts: A Comparative Study
di: Nadas, Mihai, et al.
Pubblicazione: (2025)

Synthetic Data Generation Using Large Language Models: Advances in Text and Code
di: Nadas, Mihai, et al.
Pubblicazione: (2025)

Mitigating Heterogeneous Token Overfitting in LLM Knowledge Editing
di: Liu, Tianci, et al.
Pubblicazione: (2025)

Communication Bias in Large Language Models: A Regulatory Perspective
di: Kuenzler, Adrian, et al.
Pubblicazione: (2025)

Sparse Transformer with Local and Seasonal Adaptation for Multivariate Time Series Forecasting
di: Zhang, Yifan, et al.
Pubblicazione: (2023)

Advancing continual lifelong learning in neural information retrieval: definition, dataset, framework, and empirical evaluation
di: Hou, Jingrui, et al.
Pubblicazione: (2023)

LLMic: Romanian Foundation Language Model
di: Bădoiu, Vlad-Andrei, et al.
Pubblicazione: (2025)

Data Contamination Quiz: A Tool to Detect and Estimate Contamination in Large Language Models
di: Golchin, Shahriar, et al.
Pubblicazione: (2023)

TF1-EN-3M: Three Million Synthetic Moral Fables for Training Small, Open Language Models
di: Nadas, Mihai, et al.
Pubblicazione: (2025)

The Widespread Adoption of Large Language Model-Assisted Writing Across Society
di: Liang, Weixin, et al.
Pubblicazione: (2025)

UniBERT: Adversarial Training for Language-Universal Representations
di: Avram, Andrei-Marius, et al.
Pubblicazione: (2025)