Latest Articles for Data Science

A new framework enhances knowledge graph completion efficiency and accuracy using large language models.

This article discusses uncertainty-aware models for better vector field analysis using deep neural networks.

Two specialized QA datasets aim to improve question-answering systems for Adobe Acrobat and Photoshop.

Introducing a method to assess AI models on unseen data more effectively.

Efforts to reduce confusion in AI learning from human feedback.

AIDE customizes explanations for machine learning predictions based on user intent.

Researchers enhance fetal ultrasound accuracy using synthetic images and machine learning.

A new model for classifying multi-sector companies improves investment strategies.

Explore cost-effective techniques for training text-to-image generative models.

A new dataset aims to improve facial recognition accuracy for diverse populations.

Research minimizes human labeling in reinforcement learning using concept bottleneck models.

A novel method for detecting unseen objects using self-supervised learning.

Exploring how dataset diversity impacts deep learning model performance.

New model CFGen improves synthetic single-cell data generation for better research insights.

This article discusses predicting weakly turbulent systems using limited observations and advanced methods.

A new framework enhances link polarity predictions in signed networks.

A new method boosts generative models' performance even with scarce training data.

A new model offers insights into chemical reaction mechanisms using machine learning.

Introducing Discrete Flow Matching for better discrete data generation techniques.

A method to improve language model behavior against harmful outputs.

Exploring the shift from dot products to learned similarities for better retrieval.

Learn how reduced-order models aid in analyzing complex systems using causation entropy.

A look into how dynamic graphs shape our interactions and insights.

This method enhances prompt creation for large language models, improving overall performance.

New techniques improve language understanding and processing speed in models.

A method to reduce bias in language models by making them forget harmful information.

Introducing ScholarChemQA, a dataset for chemical question answering to support researchers.

A novel method enhances discovery of dense areas in complex graphs.

AnInfoNCE improves contrastive learning by addressing challenges in data augmentation.

This research explores new methods for balancing multiple goals in reinforcement learning.

GraphTP enhances graph learning through improved data strategies.

Advancements in detecting out-of-distribution data using new techniques.

A study on balancing transfer learning and individual privacy in data usage.

A new method improves medical image analysis while protecting patient privacy.

A study on local and global approaches in continual learning algorithms.

Introducing a flexible tool for creating realistic synthetic data for recommendation systems.

QPT V2 enhances visual scoring using masked image modeling and high-quality data.

Dynamic Retraining-Updating enhances object detection without the need for labeled data.

MMTrail combines visual and audio descriptions for better video-language models.

BART is a powerful tool for predicting outcomes in various fields.