An iterative self-training method using DPO on self-generated score-conditioned explanations improves both image scoring accuracy and score-explanation consistency in VLMs.
chosen” and “rejected
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Iterative Self-Improvement of Vision Language Models for Image Scoring and Self-Explanation
An iterative self-training method using DPO on self-generated score-conditioned explanations improves both image scoring accuracy and score-explanation consistency in VLMs.