Wei Hu

A new method enhances language models through knowledge integration.

HERTA improves efficiency in training Unfolded GNNs while maintaining clarity and interpretability.

Study investigates how near-interpolating models perform on unseen data.

This paper introduces methods for better temporal fact extraction from complex sentences.

A new method enhances the ability to generate diverse texts with specific attributes.

A new method enhances code completion using data flow analysis.

Analyzing how rovers are tested on Earth before space missions.

A new model enhances inductive reasoning in knowledge graphs by using multiple starting entities.

A method enhancing machine learning to better recognize rare categories.

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

A new method enhances vehicle design for off-road conditions.

A new framework improves code generation using pair programming principles.

New findings shine light on PHEOD solitons and their applications.