LorenzoPacchiardi

Message

Two Mathematical Perspectives on AI Hallucinations and Uncertainty

A recent OpenAI preprint (and blogpost) examines the sources of AI hallucinations. This reminded me of a 2024 preprint which was similar in scope. These two papers use different mathematical lenses but arrive at complementary conclusions about the necessity of uncertainty expressions in AI systems. I briefly review and compare...

Sep 23, 20250

No Answer Needed: Predicting LLM Answer Accuracy from Question-Only Linear Probes

TLDR: This is the abstract, introduction and conclusion to the paper. See here for a summary thread. Abstract Do large language models (LLMs) anticipate when they will answer correctly? To study this, we extract activations after a question is read but before any tokens are generated, and train linear probes...

Sep 16, 202510

LESSWRONG
LW

LESSWRONG
LW

LorenzoPacchiardi

LorenzoPacchiardi

LorenzoPacchiardi

LorenzoPacchiardi

Two Mathematical Perspectives on AI Hallucinations and Uncertainty

No Answer Needed: Predicting LLM Answer Accuracy from Question-Only Linear Probes

Two Mathematical Perspectives on AI Hallucinations and Uncertainty

No Answer Needed: Predicting LLM Answer Accuracy from Question-Only Linear Probes