Jingbo Shang

This paper presents a graph-based method for enhancing patent claim approval predictions.

This study enhances relation extraction by linking it with natural language inference techniques.

Evaluating methods for precise control of text features in LLM outputs.

Examining the impact of data contamination on language model performance and evaluation.

This paper challenges the belief in self-consistency among answers from language models.

A new method for classifying text with user input and weak supervision.

Exploring the efficiency and adaptability of language models through modular design.

A new method enhances how language models follow complex instructions.