Knowledge conflicts for llms: A survey

R Xu, Z Qi, Z Guo, C Wang, H Wang, Y Zhang… - arXiv preprint arXiv …, 2024 - arxiv.org
This survey provides an in-depth analysis of knowledge conflicts for large language models
(LLMs), highlighting the complex challenges they encounter when blending contextual and …

Logical Consistency of Large Language Models in Fact-checking

B Ghosh, S Hasan, NA Arafat, A Khan - arXiv preprint arXiv:2412.16100, 2024 - arxiv.org
In recent years, large language models (LLMs) have demonstrated significant success in
performing varied natural language tasks such as language translation, question-answering …

Does Instruction Tuning Make LLMs More Consistent?

C Fierro, J Li, A Søgaard - arXiv preprint arXiv:2404.15206, 2024 - arxiv.org
The purpose of instruction tuning is enabling zero-shot performance, but instruction tuning
has also been shown to improve chain-of-thought reasoning and value alignment (Si et al …