Xiaoye Qu

This study enhances text classification accuracy using a new framework for LLMs.

New test CoTempQA improves understanding of events in language models.

Twin-Merging improves model merging efficiency and adaptability across various tasks.

A new framework aims to reduce hallucinations in LVLMs through active retrieval.

A framework to reduce false outputs in language-vision models across multiple languages.

ConflictBank offers insights into knowledge conflicts in language models.

A new approach to chatbots using a mixture of experts for efficient responses.