A new method to edit language models effectively while maintaining performance.
― 5 min read
Computer Science - Computation and Language
RSSThis study examines how language models perform language tasks similar to humans.
― 5 min read
A new dataset evaluates Large Language Models' reasoning with complex queries.
― 8 min read
Assessing question difficulty enhances the effectiveness of information retrieval systems.
― 6 min read
Explore how words change meaning over time and the factors influencing this process.
― 6 min read
Exploring new techniques for translating poetry with greater emotional depth.
― 4 min read
Explore how LLMs perform addition using unique mathematical techniques.
― 6 min read
This article examines if language models possess beliefs and follow coherence norms.
― 7 min read
A new method improves confidence scoring in language models using stable explanations.
― 9 min read
Introducing IrokoBench to improve LLM evaluation in African languages.
― 7 min read
A project aiming to simplify access to Swiss government data in multiple languages.
― 7 min read
Study explores methods for adapting AI text for language learners.
― 4 min read
Uncovering the risks posed by backdoor attacks on intelligent systems.
― 5 min read
A new framework improves pruning methods for large language models without retraining.
― 5 min read
A new dataset enhances research in linking events across documents with creative language.
― 6 min read
A study on how context affects leaderboard generation in AI research.
― 5 min read
Evaluating LLMs' strategic reasoning capabilities using diverse games.
― 7 min read
This study examines the use of AI in analyzing student answers in biology education.
― 6 min read
Examining how LLMs transform data accessibility and interaction.
― 5 min read
This article reviews how language models perform in spatial reasoning tasks.
― 7 min read
A new model replicates human-like understanding in AI systems.
― 7 min read
New methods like PromptFix help secure language models from hidden threats.
― 5 min read
Exploring multi-label classification to enhance discourse relation recognition.
― 8 min read
Evaluating methods for precise control of text features in LLM outputs.
― 13 min read
A new approach enhances language model alignment using limited human-annotated data.
― 4 min read
A new method enhances the alignment and safety of large language models.
― 6 min read
A look into how words relate within language systems over time.
― 5 min read
New method improves speech translation in noisy environments while preserving expressiveness.
― 4 min read
Circuit breakers provide a new method to prevent harmful AI outputs effectively.
― 3 min read
VISTA improves how we find information by integrating text and visuals.
― 7 min read
Explore the learning abilities of language models and their applications.
― 7 min read
ABEX uses Abstract-and-Expand to enhance training data for natural language understanding tasks.
― 8 min read
A new method to examine conversational tones in humans and AI.
― 6 min read
SPAC offers a new way to enhance language model responses.
― 6 min read
An analysis of Transformers' struggles with counting and copying tasks.
― 7 min read
MLVU benchmark aims to improve machine understanding of long videos.
― 5 min read
A new method to develop adaptable agents using diverse environments.
― 6 min read
A new method to assess commonsense reasoning in AI models through open-ended tasks.
― 8 min read
New systems enhance the classification of moral values in texts.
― 6 min read
PredEx offers predictions and explanations for legal judgments in India.
― 6 min read