Yingying Yan

Image Quality Assessment (IQA) in open-world environments remains challenging due to limited generalization and interpretability. Recent approaches based on multimodal large language models (MLLMs) introduce textual reasoning for quality prediction, yet their judgments rely heavily on semantically biased internal representations, making them insensitive to low-level perceptual degradations.

Back to Journals » International Journal of Women's Health » Volume 18 Authors Zhang W, Wang G, Yan Y, Li H, Li S, Liu L, Fan G Received 15 June 2026 Accepted for publication 29 August 2026 Published 14 September 2026 Volume 2026:18 633079 DOI https://doi.org/10.2147/IJWH.S633079 Checked for plagiarism Yes Review by Single anonymous peer review Peer reviewer comments 2 Editor who approved publication: Dr Matteo Frigerio Wenjing Zhang,1,2 Ge Wang,3 Yingying Yan,1,2 Huanhuan Li,1,2 Shuang Li,4...