New methods improve privacy and coherence using collocations in language data.
― 6 min read
Computer Science - Computation and Language
RSSA new method for rewriting text that ensures privacy and maintains meaning.
― 6 min read
A dataset to improve automated grading and feedback in engineering education.
― 6 min read
This study breaks down how transformers utilize context in language prediction.
― 9 min read
A new tool enhances Discourse Representation Theory parsing accuracy.
― 5 min read
Introducing GRASP, a benchmark for assessing spatial reasoning in language models.
― 7 min read
Exploring LLMs' effectiveness in decision-making through Dueling Bandits scenarios.
― 8 min read
Smaller open-source models offer effective solutions for automated essay and short answer scoring.
― 8 min read
Names from different countries impact how classifiers interpret social media content.
― 4 min read
Exploring how empathy enhances communication with robots and virtual assistants.
― 7 min read
Study reveals privacy risks and racial biases in Chicago police broadcasts.
― 5 min read
Analyzing how memes shape opinions through persuasive techniques.
― 4 min read
A new benchmark for assessing large language models in hypothesis testing.
― 6 min read
A framework to reduce bias in AI language models while maintaining accuracy.
― 6 min read
Evaluating methods to enhance long context performance in language models.
― 7 min read
ReGround3D improves understanding of human instructions in 3D environments.
― 4 min read
A new method improves the selection of data mixtures for language model training.
― 5 min read
A new method enhances LoRA's efficiency and effectiveness in machine learning.
― 5 min read
Exploring how synthetic data shapes machine learning models and their behavior.
― 6 min read
Simplified methods outperform complex agents in software problem-solving.
― 7 min read
DogeRM combines general and domain-specific models to enhance language model performance effectively.
― 5 min read
A new method improves user prompts for safer and more effective language model outputs.
― 4 min read
A look at Larimar's new approach to memory in language models.
― 5 min read
HyperLoader improves multi-task model training using innovative techniques and hypernetworks.
― 6 min read
A new method improves question answering in knowledge graphs using examples.
― 5 min read
Integrating graph knowledge improves performance in low-resource languages using language adapters.
― 6 min read
This article discusses methods to improve sign language translation using modern technology.
― 5 min read
A new dataset to help detect fake news in Polish online content.
― 5 min read
This study analyzes how language models adjust explanations for varying reading levels.
― 7 min read
Research shows how easily safety features can be removed from Llama 3 models.
― 5 min read
New models improve text analysis for Malaysian English using local news articles.
― 5 min read
Researchers examine methods to secure sensitive information in text classification models.
― 6 min read
A new framework enhances large model performance efficiently during fine-tuning.
― 6 min read
Study examines ChatGPT's effectiveness in explaining complex cancer reports to patients.
― 5 min read
A new framework improving predictions for large language models using historical performance data.
― 6 min read
SLIMER enhances NER performance by focusing on definitions and guidelines.
― 4 min read
Research reveals new methods for automatic diagnostic report writing using AI.
― 5 min read
This study evaluates speech technology in low-resource languages like Tunisian Arabic.
― 5 min read
Discover how retrieval-augmented generation improves information quality and response relevance.
― 5 min read
A new study assesses the understanding of economics by large language models.
― 5 min read