Xu Shen

SchemaRAG: Enhancing Knowledge-Intensive Reasoning of LLMs via Inference-Time Adaptive Schema

How Do Language Models Speak Languages? A Case Study on Unintended Code-Switching

Leveraging Submodule Linearity Enhances Task Arithmetic Performance in LLMs

Knowledge Graph Finetuning Enhances Knowledge Manipulation in Large Language Models

Deciphering and Enhancing Commonsense Reasoning in LLMs from the Perspective of Intrinsic Factual Knowledge Retrieval

Unveiling the Flow of Input-label Mappings for In-context Learning