Abstract

Both, Large Language Models (LLMs) and Knowledge Graphs (KGs) are used in various Natural Language Understanding (NLU) tasks. However, each has some benefits and disadvantages. This paper explores the pros and cons of each, and demonstrates how the two used together can help overcome some of the shortcomings. It also identifies specific applications of KG-enhanced LLMs for music-related user experiences on voice assistants. Finally, it enlists the challenges in KG-enhanced LLM applications.

References

• Roberto Navigli, Natural Language Understanding: Instructions for (Present and Future) Use, Proceedings of the Twenty Seventh International Joint Conference on Artificial Intelligence (IJCAI-18)
• A Lenci - arXiv preprint arXiv:2303.04229, 2023 - arxiv.org
• Manaal Faruqui, Dilek Hakkani-Tür, Revisiting the Boundary between ASR and NLU in the Age of Conversational Dialog Systems, Computational Linguistics (2022) 48 (1): 221–232.
• Samuel R. Bowman, George E. Dahl, What Will it Take to Fix Benchmarking in Natural Language Understanding?
• Amit Singhal. 2012. Introducing the Knowledge Graph: things, not strings. Google Blog. https://www.blog.google/products/search/introducing-knowledge-graph-things-not/
• Aidan Hogan, Eva Blomqvist, Michael Cochez, et all. Knowledge Graphs, arXiv:2003.02320
• Shaoxiong Ji, Shirui Pan, Erik Cambria, Pekka Marttinen, Philip S. Yu, A Survey on Knowledge Graphs: Representation, Acquisition and Applications. IEEE TNNLS 2021.
• Zhao, W. X. et al. (2023, March 31). A Survey of Large Language Models, https://doi.org/10.48550/arXiv.2303.18223
• Mikolov, T., Sutskever, I., Chen, K., Corrado, G. S., & Dean, J. (2013). Distributed Representations of Words and Phrases and their Compositionality. Advances in Neural Information Processing Systems, 26. https://doi.org/10.48550/arXiv.1310.4546
• https://snorkel.ai/large-language-models-llms/
• https://www.tasq.ai/blog/large-language-models/
• https://www.boost.ai/blog/a-hybrid-llm-chat-experience
• https://machinelearningmastery.com/a-gentle-introduction-to-hallucinations-in-large-language-models/
• https://spectrum.ieee.org/ai-hallucination
• Furkan Ufuk, The Role and Limitations of Large Language Models Such as ChatGPT in Clinical Settings and Medical Journalism, https://doi.org/10.1148/radiol.230276
• https://towardsdatascience.com/overcoming-the-limitations-of-large-language-models-9d4e92ad9823
• Ying Sheng, Lianmin Zheng, Binhang Yuan, Zhuohan Li, Max Ryabinin, Daniel Y. Fu, Zhiqiang Xie, Beidi Chen, Clark Barrett, Joseph E. Gonzalez, Percy Liang, Christopher Re´, Ion Stoica, Ce Zhang, FlexGen: High-Throughput Generative Inference of Large Language Models with a Single GPU, arXiv:2303.06865v2 [cs.LG] 12 Jun 2023
• https://engineb.com/2021/02/8-key-benefits-of-knowledge-graphs/
• Ciyuan Peng, Feng Xia, Mehdi Naseriparsa, and Francesco Osborne, Knowledge Graphs: Opportunities and Challenges, doi: 10.1007/s10462-023-10465-9
• Bilal Abu-Salih, Domain-specific Knowledge Graphs: A survey, King Abdullah II School of Information TechnologyThe University of Jordan
• Mahfouz, Mahmoud & Nourbakhsh, Armineh & Shah, Sameena. (2021). A Framework for Institutional Risk Identification using Knowledge Graphs and Automated News Profiling.
• Corbin L Rosset, Chenyan Xiong, Minh Phan, Xia Song, Paul N. Bennett, Saurabh Tiwary, Pretrain Knowledge-Aware Language Models, ICLR 2021 Conference
• Yu, Donghan & Zhu, Chenguang & Yang, Yiming & Zeng, Michael. (2020). JAKET: Joint Pre-training of Knowledge Graph and Language Understanding.
• Ashlesha V Kadam, “Applications of AI and NLP to advance Music Recommendations on Voice Assistants”, International Journal of Engineering, Business and Management (IJEBM), ISSN: 2456-7817 [Vol-7, Issue-3, May-Jun, 2023]
• Christophe Van Gysel, “Modeling Spoken Information eries for Virtual Assistants”, https://arxiv.org/abs/2304.13149v1