This study examines how planning aids in reducing factual errors in text generation.
― 6 min read
Computer Science - Computation and Language
RSSA new technique enhances the efficiency of pre-trained language models.
― 6 min read
A new method improves tracking of privacy leaks in large language models.
― 7 min read
Discover how AI is transforming diagnosis in computational pathology using foundation and vision-language models.
― 7 min read
PENME enhances language model updates, tackling lexical bias and computational efficiency.
― 6 min read
This approach improves summarization by accessing vast knowledge bases for relevant content.
― 7 min read
Examining how voice patterns affect meaning and technology performance.
― 4 min read
New test reveals strengths and weaknesses of LLMs in processing long texts across languages.
― 8 min read
A new automated method for creating AI research leaderboards using language models.
― 6 min read
A new approach to tailor AI systems based on individual user preferences.
― 5 min read
This study examines how LLMs perform in social deduction games through new evaluation metrics.
― 6 min read
Addressing the rise of misleading images online through better fact-checking methods.
― 7 min read
New methods improve knowledge accuracy in Vision Language Models.
― 7 min read
A new method to estimate LLM performance effectively.
― 7 min read
Easily convert PDF documents to readable formats like JSON and Markdown.
― 6 min read
A new model that enhances explanations for restaurant recommendations.
― 6 min read
A method that combines knowledge graphs with language models to improve document classification.
― 6 min read
Automation in animation creation opens new pathways for storytelling and visuals.
― 6 min read
A new method combines extractive and abstractive techniques for clearer regulatory document summaries.
― 7 min read
This paper examines how LLMs express confidence in their answers.
― 6 min read
Examining how example choices affect fairness in language models.
― 5 min read
This article discusses methods to enhance coding tools for multiple languages.
― 6 min read
A framework using advanced models to improve research literature analysis.
― 5 min read
Researchers create a dataset to improve language models' ethical decision-making.
― 7 min read
New defense method significantly reduces risks of harmful outputs in language models.
― 7 min read
A new method improves program repair by using process-based feedback.
― 6 min read
RedWhale model enhances Korean text understanding through specialized techniques.
― 6 min read
Research focuses on predicting errors in speech recognition for better accuracy.
― 5 min read
New technique for better understanding model predictions using counterfactual examples.
― 5 min read
This article examines bias in language models and strategies to reduce it.
― 5 min read
A new method enhances document classification by better detecting unfamiliar data.
― 6 min read
A new method enhances the training of small language models for better conversations.
― 6 min read
New methods enhance language models' grasp of sarcasm in communication.
― 7 min read
A method to examine political news deeply and fairly.
― 8 min read
A new method enhances theorem proving using AI and expert learning.
― 5 min read
Examining the emotional impacts of Hurricane Harvey on affected individuals and communities.
― 7 min read
Combining traditional and modern methods for improved search accuracy.
― 5 min read
FLAME improves navigation by combining language and visual input in urban settings.
― 5 min read
Examining vulnerabilities in code completion tools and their impact on user privacy.
― 5 min read
Research assesses how well LLMs generate educational questions for learning.
― 4 min read