-
MQUAKE: Assessing knowledge editing in language models via multi-hop questions
MQUAKE is a knowledge editing benchmark that includes MQUAKE-CF-3K based on counterfactual edits, and MQUAKE-T with temporal knowledge updates. -
PokeMQA: Programmable knowledge editing for Multi-hop Question Answering
Multi-hop question answering (MQA) is one of the challenging tasks to evaluate machine’s comprehension and reasoning abilities, where large language models (LLMs) have widely... -
MQUAKE-CF and MQUAKE-T datasets
The MQUAKE-CF and MQUAKE-T datasets comprise multi-hop questions that are based on real-world facts, where the edited facts are counterfactual. -
Retrieval-Augmented Knowledge Editing for Multi-Hop Question Answering in Lan...
Large Language Models (LLMs) have shown proficiency in question-answering tasks but often struggle to integrate real-time knowledge updates, leading to potentially outdated or...