Sign up to save tools and stay up to date with the latest in AI
bg
bg
1

Researchers describe how to tell if ChatGPT is confabulating

Jun 22, 2024 - arstechnica.com
Researchers from the University of Oxford have developed a method to determine when large language models (LLMs) are confabulating, or making things up. LLMs, which are trained on massive amounts of text, often provide false answers due to misinformation in their training data, inability to extrapolate from facts, or incentivization of falsehoods in their training. The researchers' method focuses on semantic entropy, which evaluates all statistically likely answers generated by the LLM and determines how many are semantically equivalent. If many answers have the same meaning, the LLM likely has the correct answer but is uncertain about phrasing. If not, the LLM is likely to confabulate.

The researchers' work is significant given the increasing reliance on LLMs for various tasks, from college essays to job applications. Their method, which works across popular models and a broad range of subjects, could help prevent LLMs from providing false information. The researchers also found that most of the "alternative facts" provided by LLMs are a product of confabulation.

Key takeaways:

  • Large Language Models (LLMs) often give false answers with confidence due to various reasons such as being trained on misinformation, inability to extrapolate from facts, or being incentivized to provide a falsehood.
  • Researchers from the University of Oxford have found a way to determine when LLMs appear to be confabulating, or making things up, a habit that is common across all popular models and a broad range of subjects.
  • LLMs are not trained for accuracy but to produce human-sounding phrasing based on the massive quantities of text they are trained on. If the training examples are few or inconsistent, LLMs synthesize a plausible-sounding but likely incorrect answer.
  • The researchers focus on semantic entropy, which evaluates all the statistically likely answers evaluated by the LLM and determines how many of them are semantically equivalent. If a large number all have the same meaning, the LLM is likely uncertain about phrasing but has the right answer. If not, it is presumably prone to confabulation.
View Full Article

Comments (0)

Be the first to comment!