Graphs and Hallucinations
Exploring the intersection of large language models and graph mining, Erwan and Gilles delve into the phenomenon of hallucinations in LLMs, particularly when generating known graphs like the Karate Club. They discuss how these hallucinations can reveal insights into the internal workings of LLMs, as they experiment with prompts to extract data and analyze the responses from multiple models. Their findings highlight the potential for understanding the capabilities and limitations of LLMs through the lens of graph data.In this clip
From this podcast
Data Skeptic
Auditing LLMs and Twitter
Related Questions