Zhixue Liang Wenyong Dong Bo Zhang CLIP-TSA: CLIP-guided open-vocabulary semantic segmentation with two-level semantic awareness. 65 2025 February 31 Multim. Syst. 1 https://doi.org/10.1007/s00530-024-01666-5 db/journals/mms/mms31.html#LiangDZ25 streams/journals/mms