AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Data Skeptic
Auditing LLMs and Twitter
Jan 29, 2025
In this insightful discussion, Erwan Le Merrer, a collaborator in graph theory and distributed systems, reveals how graph-based techniques can expose patterns in large language models and shadow banning on Twitter. He explains the application of epidemic models to examine shadow banning spread across user networks. The conversation also highlights the use of graph metrics to audit LLM outputs, the challenge of bias detection in AI, and innovative methodologies for understanding algorithmic behavior, shedding light on often-overlooked platform moderation practices.
40:26
Episode guests
AI Summary
AI Chapters
Episode notes
Podcast summary created with Snipd AI
Quick takeaways
- Graph theory can help audit LLMs by revealing inaccuracies in model outputs and providing insights into training data structures.
- The analysis of shadow banning on Twitter suggests targeted actions through algorithms rather than random software errors, highlighting network behavior complexities.
Deep dives
Intersection of LLMs and Graphs
The podcast discusses the challenge of exploring large language models (LLMs) and their relationship with graph data structures. In particular, it addresses how LLMs, primarily designed for text prediction, attempt to reproduce data from classic graphs like the Karate Club dataset. An example highlighting this was when the hosts tested ChatGPT's ability to generate the edge list from the Karate Club dataset, revealing that it inaccurately omitted some edges. This led to a broader discussion on how LLMs manage graph structures, suggesting that their limitations in graph representation stem from their core functionality in text processing.
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode