Latest Articles for Fine-tuning

Examining the impact of synthetic news content and detection difficulties.

Examining memorization in code completion models and its privacy implications.

This article examines ways to improve planning abilities in large language models.

A method to evaluate model knowledge through internal processing.

DetectBench evaluates LLMs on their ability to detect hidden evidence in reasoning tasks.

A new method to improve model stability and performance in low-resource settings.

How fine-tuning affects language models' ability to recall facts accurately.

Advancements in fine-tuning language models using innovative techniques.

RankAdaptor optimizes fine-tuning for pruned AI models, enhancing performance efficiently.

Methods to reduce memory usage during fine-tuning of large models.

This study presents a dataset and method to enhance Chinese ASR accuracy using Pinyin.

New methods refine reasoning skills in language models for better task performance.

A new method enhances how language models align with human values.

This study focuses on enhancing model responses by targeting specific length requirements.

Research on improving knowledge transfer in resource-limited smart devices.

This study evaluates how well large language models use external information.

GTZAN-synth dataset leverages synthetic music for better music tagging systems.

New method enhances spiking neural networks' performance in language tasks.

New methods improve molecular design by measuring prediction uncertainty.

A new system enhances data processing while ensuring user privacy and efficient resource use.

HyperLoader improves multi-task model training using innovative techniques and hypernetworks.

Research shows how easily safety features can be removed from Llama 3 models.

A new framework enhances large model performance efficiently during fine-tuning.

CPT improves black-box model performance without direct access to internal parameters.

Fine-tuning large language models directly on smartphones while protecting user data.

Examining methods to enhance code generation for specialized programming languages using LLMs.

New dataset enhances Arabic language model performance and fosters effective communication.

Techniques to reduce harmful language generation in AI models.

A method to enhance language models by creating engaging multi-turn dialogues.

A new method to improve model performance on out-of-distribution data.

A novel method to fine-tune language models efficiently with fewer parameters.

This study examines Mix-Training for keyword spotting in noisy speech conditions.

CLIP-CITE enhances CLIP models for specialized tasks while retaining flexibility.

A new method improves image generation using limited datasets effectively.

Improving speech recognition systems for languages with limited online data.

Explore the advantages and applications of Low-Rank Adaptation in AI models.

A new method improves NLP models by focusing on syntactic transformations.

This study focuses on reducing gender bias in AI language models through inclusive language.

Machines improve at answering questions about images through structured training.

This article explores overparameterization and its impact on model training efficiency.