Fine-tuned vision-language models make more correct predictions, but a larger share of those correct predictions are based on invalid evidence than zero-shot models.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Beyond Accuracy: On the Effects of Fine-tuning Towards Vision-Language Model's Prediction Rationality
Fine-tuned vision-language models make more correct predictions, but a larger share of those correct predictions are based on invalid evidence than zero-shot models.