LookAlike mines inconsistent model generations as negative preference pairs and alternates SFT with DPO to improve error-distractor consistency in math MCQ generation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LookAlike: Consistent Distractor Generation in Math MCQs
LookAlike mines inconsistent model generations as negative preference pairs and alternates SFT with DPO to improve error-distractor consistency in math MCQ generation.