5 papers
Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs
Yanyan Luo, Xue Han, Ruiqiao Bai +10
Large Language Models (LLMs) have enabled increasingly personalized interactions by adapting to users' preferences, contexts, and long-term histories. However, the mechanisms that…
Understanding New-Knowledge-Induced Factual Hallucinations in LLMs: Analysis and Interpretation
Renfei Dang, Peng Hu, Zhejian Lai +3
Prior works have shown that fine-tuning on new knowledge can induce factual hallucinations in large language models (LLMs), leading to incorrect outputs when evaluated on previousl…
Large Language Models Are Cross-Lingual Knowledge-Free Reasoners
Peng Hu, Sizhe Liu, Changjiang Gao +5
Large Language Models have demonstrated impressive reasoning capabilities across multiple languages. However, the relationship between capabilities in different languages is less e…
Large Language Models are Limited in Out-of-Context Knowledge Reasoning
Peng Hu, Changjiang Gao, Ruiqi Gao +2
Large Language Models (LLMs) possess extensive knowledge and strong capabilities in performing in-context reasoning. However, previous work challenges their out-of-context reasonin…
Multilingual Pretraining and Instruction Tuning Improve Cross-Lingual Knowledge Alignment, But Only Shallowly
Changjiang Gao, Hongda Hu, Peng Hu +3
Despite their strong ability to retrieve knowledge in English, current large language models show imbalance abilities in different languages. Two approaches are proposed to address…