Verifiable and Navigable LLMs with Knowledge Graphs

Graphs, especially knowledge graphs, are powerful tools for structuring data into interconnected networks. The structured format of knowledge graphs enhances the performance of LLM-based systems by improving information retrieval and ensuring the use of reliable sources. By integrating knowledge graphs in LLM-based applications, we can reduce reliance on purely vector-based methods, which may not consistently generate accurate outputs. This integration can lead to more reliable and trustworthy results. In this talk, we will demonstrate how knowledge graphs enhance factual accuracy in responses and how their relationship-driven features enable LLM-based systems to generate more contextually-aware outputs. We will present real-life examples with side-by-side comparisons to illustrate these benefits.

Main Takeaways:
  1. Richer Contexts for Better Answers: Knowledge graphs link data in ways that provide LLMs with deeper context. This helps LLMs find and generate more relevant responses.

  2. Transparent Information Paths: Using both vector search and knowledge graphs in LLMs not only increases the precision of answers but also lets users see where information comes from, building trust in the system.

  3. Broad Use Cases: Knowledge graphs, as organized data structures, are widely applicable across many fields, including healthcare, e-commerce, HR, etc. These graphs organize and provide access to a vast amount of structured data, effectively mapping relationships between data points. This allows for a nuanced understanding and navigation of complex information. When integrated with LLMs, knowledge graphs enable them to leverage this structure to generate human-like responses that are both accurate and contextually relevant. This integration not only improves the effectiveness of data retrieval but also enhances the quality of interactions and decision-making across various applications.


Speaker

Leann Chen

AI Developer Advocate @Diffbot

Leann is a Generative AI Developer Advocate at Diffbot, who currently focuses on enhancing the performance of LLM-based applications by integrating the strengths of knowledge graphs.

Read more

From the same track

Session

No More Spray and Pray— Let's Talk About LLM Evals

The pace of development in AI in the past year or so has been dizzying, to say the least, with new models and techniques emerging weekly. Yet, amidst the hype, a sobering reality emerges: much of these advancements lack robust empirical evidence.

Speaker image - Apoorva Joshi

Apoorva Joshi

AI Developer Advocate @MongoDB

Session

Recommender and Search Ranking Systems in Large Scale Real World Applications

Recommendation and search systems are two of the key applications of machine learning models in industry. Current state of the art approaches have evolved from tree based ensembles models to large deep learning models within the last few years.

Speaker image - Moumita Bhattacharya

Moumita Bhattacharya

Senior Research Scientist @Netflix

Session

Why Most Machine Learning Projects Fail to Reach Production and How to Beat the Odds

Despite the hype around AI, many ML projects fail, with only 15% of businesses' ML projects succeeding, according to McKinsey. Particularly with the significant investments in large language models and generative AI, only a small portion of companies have managed to realize their true value.

Speaker image - Wenjie Zi

Wenjie Zi

Senior Machine Learning Engineer and Tech Lead @Grammarly