2000 character limit reached
Prompt-Time Symbolic Knowledge Capture with Large Language Models (2402.00414v1)
Published 1 Feb 2024 in cs.CL and cs.AI
Abstract: Augmenting LLMs with user-specific knowledge is crucial for real-world applications, such as personal AI assistants. However, LLMs inherently lack mechanisms for prompt-driven knowledge capture. This paper investigates utilizing the existing LLM capabilities to enable prompt-driven knowledge capture, with a particular emphasis on knowledge graphs. We address this challenge by focusing on prompt-to-triple (P2T) generation. We explore three methods: zero-shot prompting, few-shot prompting, and fine-tuning, and then assess their performance via a specialized synthetic dataset. Our code and datasets are publicly available at https://github.com/HaltiaAI/paper-PTSKC.
- Large Language Models and Knowledge Graphs: Opportunities and Challenges, August 2023a. URL http://arxiv.org/abs/2308.06374. arXiv:2308.06374 [cs].
- Unifying Large Language Models and Knowledge Graphs: A Roadmap, June 2023b. URL http://arxiv.org/abs/2306.08302. arXiv:2306.08302 [cs].
- CycleGT: Unsupervised Graph-to-Text and Text-to-Graph Generation via Cycle Training, December 2020. URL http://arxiv.org/abs/2006.04702. arXiv:2006.04702 [cs].
- INFINITY: A Simple Yet Effective Unsupervised Framework for Graph-Text Mutual Conversion, September 2022. URL http://arxiv.org/abs/2209.10754. arXiv:2209.10754 [cs].
- Entity-Relation Extraction as Multi-Turn Question Answering, September 2019. URL http://arxiv.org/abs/1905.05529. arXiv:1905.05529 [cs].
- Language Models are Few-Shot Learners, July 2020. URL http://arxiv.org/abs/2005.14165. arXiv:2005.14165 [cs].
- Large Language Model Is Not a Good Few-shot Information Extractor, but a Good Reranker for Hard Samples!, October 2023. URL http://arxiv.org/abs/2303.08559. arXiv:2303.08559 [cs].
- LLMs for Knowledge Graph Construction and Reasoning: Recent Capabilities and Future Opportunities, May 2023. URL http://arxiv.org/abs/2305.13168. arXiv:2305.13168 [cs].
- MLX: Efficient and flexible machine learning on apple silicon. https://github.com/ml-explore, 2023.
- QLoRA: Efficient Finetuning of Quantized LLMs, May 2023. URL http://arxiv.org/abs/2305.14314. arXiv:2305.14314 [cs].
- Open llm leaderboard. https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard, 2023.
- Mistral 7B, October 2023. URL http://arxiv.org/abs/2310.06825. arXiv:2310.06825 [cs].
- Llama 2: Open foundation and fine-tuned chat models, 2023.
- A Survey of Data Augmentation Approaches for NLP, December 2021. URL http://arxiv.org/abs/2105.03075. arXiv:2105.03075 [cs].
- Data Augmentation Approaches in Natural Language Processing: A Survey. AI Open, 3:71–90, 2022. ISSN 26666510. doi: 10.1016/j.aiopen.2022.03.001. URL http://arxiv.org/abs/2110.01852. arXiv:2110.01852 [cs].
Collections
Sign up for free to add this paper to one or more collections.