Person-fit statistics derived from item response theory separate human and chatbot multiple choice response patterns at low pollution levels, but the signal weakens as AI answers become common.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Applying IRT to Distinguish Between Human and Generative AI Responses to Multiple-Choice Assessments
Person-fit statistics derived from item response theory separate human and chatbot multiple choice response patterns at low pollution levels, but the signal weakens as AI answers become common.