glossary termAngestromTrust 60Published 3mo agoLive · 2mo ago
Hallucination
When a model states something fluent but false.
When a model states something fluent but false. When a model states something fluent but false.
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- LinkedLinked via unknownRefusal as a single linear direction →
- LinkedLinked via unknownHallucination in World Models is Predictable and Preventable →
- PossiblePossibly related (embedding) · 71%Models Produce Hallucinations Because of Probabilistic Training - Let's Data Science →
- PossiblePossibly related (embedding) · 65%Most RAG Hallucinations Are Retrieval Failures: How the Retrieval Brick Decides What the Model Can Invent - Towards Data Science →
- PossiblePossibly related (embedding) · 47%Most RAG Hallucinations Are Extraction Errors: Seven Patterns for a Typed Generation Contract - Towards Data Science →
- PossiblePossibly related (embedding) · 50%cvs-health/uqlm →
Related to
Related to (incoming)
Covers (incoming)
newsModels Produce Hallucinations Because of Probabilistic Training - Let's Data SciencenewsMost RAG Hallucinations Are Retrieval Failures: How the Retrieval Brick Decides What the Model Can Invent - Towards Data SciencenewsMost RAG Hallucinations Are Extraction Errors: Seven Patterns for a Typed Generation Contract - Towards Data Science
Related across the graph
repocvs-health/uqlmnewsMost RAG Hallucinations Are Retrieval Failures: How the Retrieval Brick Decides What the Model Can Invent - Towards Data SciencenewsModels Produce Hallucinations Because of Probabilistic Training - Let's Data SciencenewsMost RAG Hallucinations Are Extraction Errors: Seven Patterns for a Typed Generation Contract - Towards Data SciencepaperRefusal as a single linear directionpaperHallucination in World Models is Predictable and Preventable