Latest Articles for Training Data

Zamba is a hybrid language model combining state-space and transformer architectures.

A method for generating quality training data for language model fine-tuning.

A new framework focuses on enhancing dataset quality for better recommendations.

This article examines how learning theories tackle distribution changes.

Exploring how quantum computing influences machine learning techniques.

This article discusses challenges in few-shot fine-tuning of diffusion models and solutions.

This article discusses using smaller models to refine training data for better performance.

Investigating how small errors in training data enhance AI-generated content.

New techniques improve quality and training for 3D images.

Analyzing how AI learns from data reveals significant gaps in logic and reasoning.

A framework to identify and reduce biases in training datasets.

RoboCasa simulates environments for robots to learn everyday tasks effectively.

Exploring how LLMs use reasoning to tackle complex tasks.

Explore the learning abilities of language models and their applications.

Innovative strategies enhance material predictions using machine learning surrogates.

Testing LLMs is essential for safe and effective AI applications.

A new method improves translating mixed-language speech into English.

A new method uses natural language explanations to improve entity matching.

Introducing reflective augmentation to improve language models’ math problem-solving skills.

Examining the roots and implications of bias in language technology.

A new model enhances news article suggestions across multiple languages.

New algorithms enhance predictions of quantum ground states using limited data.

UrbanLLM streamlines urban management by breaking down complex city-related queries.

Two new models aim to improve technology access for Galician speakers.

Examining how LLMs exhibit personality traits through new testing methods.

Data contamination affects the evaluation of large language models significantly.

A new method improves privacy protection in language models while maintaining performance.

ATLAS enhances seismic data selection using active learning and representation shifts.

A simple method to create voices and control emotions in speech synthesis.

A new technique improves imaging of brain blood vessels, aiding research.

An analysis of language models and their role in healthcare.

A flexible approach for generating CFEs that respects data privacy concerns.

Exploring fairness issues in AI language models and their implications.

A new method enhances accuracy in question-answering for black-box language models.

This article analyzes repetitive structures in text generated by language models.

Insights on the challenges of machine learning in predicting material properties.

A new method uses translation to enhance language model training.

A new approach enhances reasoning in language models by generating controlled errors.

This study breaks down how transformers utilize context in language prediction.

Code poisoning enhances risks of membership inference attacks on sensitive data.