ResearchBenchmarksSemantic Segmentation on ImageNet curated subset (val)For AI assistantsMarkdownContent indexSemantic Segmentation on ImageNet curated subset (val)Follow77.55Pixel AccuracyCLIP-SVD77.227677.311377.39577.4787Sep 3, 2025Evaluation resultsMethodMethodLinksPixel AccuracymIoUmAPCLIP-SVDFramework=TextSpan, St...Framework=TextSpan, Status=fine-tuned2025.0977.5558.2782.82CLIPFramework=TextSpan, St...Framework=TextSpan, Status=pre-trained2025.0977.2457.7382.62