Language Integration in Fine-Tuning Multimodal Large Language Models for Image-Based Regression

Benchmark Model Rank Results
aesthetics-quality-assessment-on-aesthetic-visual-analysisRvTC (image-only)–SRCC: 0.833PLCC: 0.831
aesthetics-quality-assessment-on-aesthetic-visual-analysisRvTC+–SRCC: 0.899PLCC: 0.901
no-reference-image-quality-assessment-on-kadid-10kRvTC (image-only)–SRCC: 0.98PLCC: 0.98