The paper addresses the challenges of short-form UGC video quality assessment, particularly the ambiguous content and complex distortions introduced by advanced creation modes and sophisticated processing workflows. To tackle these issues, the authors establish the first large-scale kaleidoscope short-form video database, named KVQ, which includes 4200 user-uploaded and processed videos. The database is annotated with absolute quality scores and partial ranking scores for indistinguishable samples by professional researchers. Based on this database, the authors propose the first short-form video quality evaluator, KSVQE, which incorporates content understanding using CLIP and distortion understanding using CONTRIQUE. Experimental results show that KSVQE outperforms existing methods on the KVQ dataset and popular VQA datasets, demonstrating its effectiveness in handling complex short-form videos. The project is available at <https://lixinustc.github.io/projects/KVQ/>.The paper addresses the challenges of short-form UGC video quality assessment, particularly the ambiguous content and complex distortions introduced by advanced creation modes and sophisticated processing workflows. To tackle these issues, the authors establish the first large-scale kaleidoscope short-form video database, named KVQ, which includes 4200 user-uploaded and processed videos. The database is annotated with absolute quality scores and partial ranking scores for indistinguishable samples by professional researchers. Based on this database, the authors propose the first short-form video quality evaluator, KSVQE, which incorporates content understanding using CLIP and distortion understanding using CONTRIQUE. Experimental results show that KSVQE outperforms existing methods on the KVQ dataset and popular VQA datasets, demonstrating its effectiveness in handling complex short-form videos. The project is available at <https://lixinustc.github.io/projects/KVQ/>.