Large language models (LLMs) such as GPT-3 have shown remarkable capabilities in transforming various natural language processing (NLP) tasks across different domains. However, LLMs often generate incorrect answers, known as hallucinations, posing significant challenges to their usability and reliability. Additionally, LLMs operate as black boxes, making it difficult to understand how they arrive at specific conclusions, leading to transparency and explainability issues. Combining LLMs with KGs creates a powerful synergy that significantly enhances the capabilities of artificial intelligence across various tasks. This integration leverages the strengths of both technologies, with LLMs excelling at understanding and generating human-like text, and KGs providing structured, reliable information about entities and their relationships. Together, they offer a robust approach to problem-solving across diverse domains.

This seminar will focus on the intersection of LLMs and KGs, covering areas of interest including, but not limited to:

• KG completion using LLMs
• Question answering with KGs and LLMs
• Explainability of LLMs with KG integration
• Reasoning with LLMs and KGs
• Enhanced prompt engineering using KGs

Contributions of the students:

Each student will be assigned one paper on the topic, which could be a research paper discussing a novel approach or a resource paper presenting datasets, tools, etc. The student will be responsible for the following tasks:

1. Report Writing: Read the assigned paper thoroughly and write a 15-page seminar report explaining the methods and findings in their own words.
2. Presenting: Prepare and deliver a seminar presentation to share insights from the paper with other seminar participants.
3. Conducting Experiments: If the authors provide code, re-implement it for small-scale experiments using Google Colab or make the implementation available via GitHub.