Latest Articles for Performance

An analysis of language models and their role in healthcare.

This study evaluates how well LLMs reason about cardinal directions.

A new loss function, Astral, boosts the performance of physics-informed neural networks.

This study benchmarks Language Models' performance using Italian INVALSI tests.

Examining how hyper-parameters shape the effectiveness of deep RL agents.

Study shows LLMs favor native speakers, raising equality concerns.

CharED combines language models for improved performance without shared vocabularies.

New methods improve summarizing long texts across various fields.

Learn about the Quantum Application Score for assessing quantum device capabilities.

Automatic tracing simplifies task management in parallel programming for better performance.

Task fusion enhances performance in distributed computing through efficient task management.

CompassDB offers a fast and efficient key-value storage solution.

New algorithms improve online linear regression for dynamic environments.

A new adaptive optimizer reduces memory use in training large neural networks.

This study examines how explanations impact user perceptions of AI capabilities.

A look at how companies in Europe are rated on sustainability.

A new method boosts matrix multiplication performance using sparse matrices.

New benchmark improves evaluation of multimodal models by minimizing biases.

Research shows tuning with English data may enhance multilingual information retrieval.

This work focuses on better number representation using digit embeddings for improved predictions.

Study examines learned indexes' performance for joins with data stored externally.

CRAB enhances testing for language models in real-world environments.

A new method improves the selection of data mixtures for language model training.

Research shows how easily safety features can be removed from Llama 3 models.

M-SAT enhances Decision Transformers for complex action tasks.

A method to enhance timbre in music production through synthesizers.

CPT improves black-box model performance without direct access to internal parameters.

A study compares Large Language Models and top human authors in creative writing.

This study examines how large language models handle fuzzy reasoning tasks.

This article examines how small language models learn to handle noise in data.

Investigating how overparameterized models outperform underparameterized ones in learning features.

Learn how power meters impact cycling performance and training methods.

New methods improve model merging for better machine learning performance.

Quadrupedal robots are learning to walk on two legs, improving versatility.

This study examines privacy issues and protection methods for AI classifiers.

A new method improves reinforcement learning efficiency and performance using physical models.

Exploring optimization strategies for the unbounded knapsack problem in online settings.

A new approach to improve efficiency and speed in language models.

Studying how quantization affects performance in different languages.

DCoT enhances language model performance through multiple reasoning paths.