Papers by Yanxu Zhu

1 papers
KG-FPQ: Evaluating Factuality Hallucination in LLMs with Knowledge Graph-based False Premise Questions (2025.coling-main)

Copied to clipboard

Challenge: Existing benchmarks that assess this vulnerability rely on manual construction, resulting in limited size and lack of expandability.
Approach: They propose a method to generate false premise questions based on knowledge graphs . they modify true triplets extracted from KGs to create false premises .
Outcome: The proposed method generates semantically rich FPQs using state-of-the-art GPTs.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations