Across five OpenAI models, mentioning a correct answer in a query boosts LLM accuracy by up to 15 points, while mentioning an incorrect answer lowers it by a similar amount.
Okun, and Emily Oster
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
"Check My Work?": Measuring Sycophancy in a Simulated Educational Context
Across five OpenAI models, mentioning a correct answer in a query boosts LLM accuracy by up to 15 points, while mentioning an incorrect answer lowers it by a similar amount.