Large language models (LLMs), such as ChatGPT, are able to generate human-like, fluent responses for many downstream tasks, e.g., task-oriented dialog and question answering. However, applying LLMs to real-world, mission-critical applications remains challenging mainly due to their tendency to generate hallucinations and their inability to use external knowledge. This paper proposes a LLM-Augmenter system, which augments a black-box LLM with a set of plug-and-play modules. Our system makes the LLM generate responses grounded in external knowledge, e.g., stored in task-specific databases. It also iteratively revises LLM prompts to improve model responses using feedback generated by utility functions, e.g., the factuality score of a LLM-generated response. The effectiveness of LLM-Augmenter is empirically validated on two types of scenarios, task-oriented dialog and open-domain question answering. LLM-Augmenter significantly reduces ChatGPT's hallucinations without sacrificing the fluency and informativeness of its responses. We make the source code and models publicly available.
Think Carefully and Check Again! Meta-Generation Unlocking LLMs for Low-Resource Cross-Lingual Summarization
University of California, San Diego(加利福尼亚大学圣迭戈分校);Nanyang Technological University(南洋理工大学);National University of Singapore(新加坡国立大学);Tsinghua University(清华大学)
May I Check Again? -- A simple but efficient way to generate and use contextual dictionaries for Named Entity Recognition. Application to French Legal Texts
ChartCheck: Explainable Fact-Checking over Real-World Chart Images
King’s College London(伦敦国王学院);University of Utah(犹他大学);University of Pennsylvania(宾夕法尼亚大学);TIB – Leibniz Information Centre for Science and Technology(莱布尼茨科学与技术信息中心(TIB))
SciCom Wiki: Fact-Checking and FAIR Knowledge Distribution for Scientific Videos and Podcasts
L3S Research Center, Leibniz University Hannover(莱布尼茨汉诺威大学L3S研究中心);Cluster of Excellence SE²A – Sustainable and Energy-Efficient Aviation(卓越中心SE²A——可持续和节能航空);TIB - Leibniz Information Centre for Science and Technology(莱布尼茨科学与技术信息中心)
本文提出以 Wikibase 为核心的 SciCom Wiki 平台,并结合神经符号事实核查与知识图谱,支持科学视频和播客的 FAIR 表示、协作策展及内容核验。