Latest Articles for Machine Learning

New model improves coreference resolution through parallel multilingual data.

A new approach to improve anomaly detection using k-Nearest Neighbors of Neighbors.

A new method leverages machine learning for quick supernova classification.

CardIndex combines indexing and cardinality estimation for faster, resource-efficient data handling.

A new method uses deep learning to enhance fluid dynamics research efficiency.

A novel approach reduces document processing time and resources for language models.

Caterpillar is a novel MLP architecture for capturing local image details.

This article presents tools to interpret the training of error-correcting codes.

A look at how differential privacy safeguards individual information in data analysis.

A new approach for running X3D model on FPGAs for efficient video analysis.

This paper discusses methods to enhance communication efficiency in split learning while preserving model performance.

A framework to enhance generative models using pre-trained diffusion models.

This article examines how to detect inaccuracies in language model-generated references.

A new approach integrates kernel methods with deep learning for better performance.

Research develops sentence collection for identifying subjectivity in news reporting.

A new method enhances vision-language models through real-time feedback for better performance.

A new approach enhances language models by directly optimizing for human preferences.

Research reveals how brains simulate future events using complex models.

DeepVAT improves clustering assessment in complex image datasets using self-supervised learning.

A new model approach for better data predictions using Squared Neural Families.

BioCLIP combines protein sequences and structures for better understanding.

Exploring self-repellent random walks to improve sampling efficiency in complex systems.

LayoutMask enhances text and layout interaction for better document comprehension.

A look into weighted classification metrics and score-oriented losses in neural networks.

A new approach to computing that prioritizes energy savings over strict accuracy.

A new model improves accuracy in predicting traffic conditions.

This research focuses on using keywords to enhance language model responses.

A new method to train temporal knowledge graphs without forgetting past information.

A look at differentially private convex optimization for data protection in machine learning.

New methods enhance efficiency in analyzing large tensors with randomized algorithms.

A new model enhances confidence scores in speech recognition systems.

A new approach to improve scene graph generation for better visual understanding.

UMD offers a new way to identify complex backdoor attacks effectively.

Research shows methods to better assign bug reports to developers.

Boltzmann generators offer new insights into complex systems and phase transitions.

This article discusses the role of Deep Equilibrium models in federated learning.

Combining graph rewriting with GNNs for dynamic data analysis.

Exploring the Deep Unconstrained Features Model and its impact on neural networks.

W-Procer enhances named entity recognition in medical texts using few-shot learning.

New algorithms improve decoding in Hierarchical Hidden Markov Models.