Safety & Red Teaming

Adversarial testing, toxicity evaluation, and safety assessment methods for ensuring AI systems behave within acceptable boundaries.

Where to Start

This cluster covers 1 topic. Here's a suggested reading order from fundamentals to advanced.

1

Hallucination

Start here

Hallucination is what happens when a large language model generates text that sounds confident and coherent but is factually wrong or entirely fabricated. It stems from the statistical nature of next-token prediction, where the model optimizes for plausibility rather than truth. Detection, mitigation through grounding and retrieval, and careful system design are active areas of research and engineering practice. Also known as: LLM Hallucination, AI Hallucination

6 articles

Explore by Perspective

Related Themes

Embeddings & Vector Search 1 topics Inference Optimization 1 topics LLM Training & Pre-Training 1 topics Model Evaluation & Benchmarks 1 topics Transformer & Attention Internals 1 topics