Latest Articles for Hallucinations

This article examines issues of code hallucination in LLMs and their implications.

A study on hallucinations in language models and their training implications.

New tools enhance surgical training by utilizing video and text data.

This study examines how planning aids in reducing factual errors in text generation.

A method to enhance accuracy in language models by detecting hallucinations.

ConVis aims to minimize inaccuracies in multimodal large language models.

A new framework improves answer accuracy in AI models by focusing on evidence.

Introducing LRP4RAG, a method to better detect hallucinations in language models.

A new framework enhances image captioning accuracy and reduces errors.

A new method for detecting hallucinations in language models using corrupted data.

A look at the complexities and improvements in speech-to-speech translation technology.

A new method to assess uncertainty in language model outputs for greater reliability.

A new method addresses the challenge of detecting inaccuracies in AI-generated text.

A new model improves how researchers manage citations and scholarly articles.

A new framework enhances text descriptions using images and structured data.

A new approach to reduce inaccuracies in language models using skepticism.

Examining the risks and challenges of AI technology in medical applications.

Study reveals how language models utilize context for accurate responses.

Examining AI hallucinations and critical thinking in human interactions.

A new framework enhances accuracy and reduces errors in medical text generation.

THaMES offers a framework to reduce hallucinations in language models.

New methods aim to reduce inaccuracies in language models within information retrieval systems.

This research tests a tool to improve accuracy of traffic-based language models.

A new method improves detection of inaccuracies in language models.

SLaVA-CXR improves chest X-ray report generation for better clinical efficiency.

PACU framework enhances VLLMs by refining prompts and utilizing image captions.

Examining the accuracy issues in large language models and their societal effects.

A new framework improves detection of false outputs in language models using unlabeled data.

AI tools in healthcare offer benefits but raise significant safety concerns.

Assessing the impact of LLMs on healthcare documentation and safety.

Exploring how large language models can mislead users in medical advice.

Exploring how smaller models struggle with inaccuracies from larger counterparts.

RadFlag helps ensure AI-generated medical reports are accurate and trustworthy.

New method improves accuracy in vision-language models by reducing hallucination.

Research shows ways to enhance context awareness in language models for better responses.

A new dataset evaluates large language models for predicting material properties.

A look into new methods for enhancing trust in AI responses.

AI programs create personalities through interactions, forming unique identities in a virtual space.

New tool H-POPE improves accuracy of vision-language models.

New methods improve the accuracy of large language models.