Latest Articles for Language Models

AutoHint improves prompt quality for better language model performance.

Introducing a framework for advanced text generation using language models.

Gender-tuning reduces bias in language models while maintaining performance levels.

Research shows illogical prompts can yield strong results in language models.

This research shows how GPT-4 can improve image classification accuracy using visual descriptions.

New research highlights the importance of ripple effects in updating language models.

A method that improves adaptability and efficiency in language models.

A new method boosts word meaning clarity across languages.

A study comparing in-context learning and supervised learning reveals key differences in model performance.

Camoscio is a new language model tailored for Italian users, enhancing research opportunities.

A new test suite evaluates safety behaviors in language models.

A study on using prompt chaining to enhance legal document classification.

A new approach improves simile detection through data augmentation and language modeling.

A look into distinguishing AI-generated text from human writing.

A novel approach improves logical abilities in language models through structured thinking.

A new benchmark for evaluating language models in Chinese medical contexts.

A new benchmark aims to reduce false information in healthcare applications of LLMs.

New methods combine fast and slow reasoning for improved visual problem-solving.

This article examines the performance of various language models in key NLP tasks.

SciEval evaluates language models on their scientific research skills with diverse questioning.

This study explores how ChatGPT can improve open intent detection.

A study compares human and GPT model color perceptions.

Research shows data cleaning is crucial for enhancing language model effectiveness.

A method to improve language model use in business settings.

A new framework boosts large language models' ability to use tools effectively.

A specialized language model designed to enhance radiology reporting and efficiency.

This paper explores the extraction of terms from mathematical texts using ChatGPT.

A critical look at GPT-3's ability to solve analogy problems.

New methods enhance memory use and speed in language model training.

HAE-RAE Bench focuses on assessing cultural knowledge in Korean language models.

A new approach improves language model performance through optimized weight rounding.

phi-1.5 shows that smaller models can perform well with quality data.

This research evaluates how language models meet readability standards in educational content.

A new model improves understanding of code-mixed languages.

Study analyzes fine-tuning methods for language models to retain knowledge across languages.

This study assesses how LLMs can be used in real-time translation tasks.

This study evaluates when expansions improve or harm information retrieval performance.

Exploring how attackers exploit large language models for knowledge extraction.

KoBigBird-large enhances understanding of Korean with improved text processing.

New encoder-decoder models enhance Spanish language processing capabilities.