LLMs implicitly plan answer positions during MCQ generation, as shown by predictive signals in hidden representations and controllable shifts via activation steering.
Addressing Blind Guessing: Calibration of Selection Bias in Multiple-Choice Question Answering by Video Language Models , ISBN=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Do Large Language Models Plan Answer Positions? Position Bias in Multiple-Choice Question Generation
LLMs implicitly plan answer positions during MCQ generation, as shown by predictive signals in hidden representations and controllable shifts via activation steering.