[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-1-en-105":3,"doc-seo-194473-105":53,"doc-detail-194473-en":126},{"code":4,"msg":5,"data":6},0,"success",[7,14,19,24,29,34,39,44,49],{"id":8,"doc_module":9,"doc_module_name":10,"category_name":11,"show_sort_weight":12,"slug":13},11,1,"Template","Presentations",90,"presentations",{"id":15,"doc_module":9,"doc_module_name":10,"category_name":16,"show_sort_weight":17,"slug":18},12,"Resumes",80,"resumes",{"id":20,"doc_module":9,"doc_module_name":10,"category_name":21,"show_sort_weight":22,"slug":23},14,"Invoices",70,"invoices",{"id":25,"doc_module":9,"doc_module_name":10,"category_name":26,"show_sort_weight":27,"slug":28},15,"Posters",60,"posters",{"id":30,"doc_module":9,"doc_module_name":10,"category_name":31,"show_sort_weight":32,"slug":33},16,"Social Media",50,"social-media",{"id":35,"doc_module":9,"doc_module_name":10,"category_name":36,"show_sort_weight":37,"slug":38},17,"Forms",40,"forms",{"id":40,"doc_module":9,"doc_module_name":10,"category_name":41,"show_sort_weight":42,"slug":43},18,"Letters",30,"letters",{"id":45,"doc_module":9,"doc_module_name":10,"category_name":46,"show_sort_weight":47,"slug":48},21,"Paper Templates",5,"papers-templates",{"id":50,"doc_module":9,"doc_module_name":10,"category_name":51,"show_sort_weight":4,"slug":52},158,"General","general-158",{"code":4,"msg":54,"data":55},"ok",{"site_id":56,"language":57,"slug":58,"title":59,"keywords":60,"description":61,"schema_data":62,"social_meta":119,"head_meta":121,"extra_data":123,"updated_unix":125},105,"en","tfsr20250147-performance-metrics-table-for-aqualora-and-stable-diffusion","tfsr20250147 - performance metrics table for AquaLoRA and Stable Diffusion","","Quantitative results compare multiple reconstruction or attack/evasion methods on image or diffusion-related experiments, reporting ACC, TPR@1%FPR, PSNR, and SSIM across several noise or perturbation settings. Tables evaluate AquaLoRA and Stable Diffusion variants (v1.4, v2.0, v2.1), include baselines such as Original and JPEG, and test conditions like Brightness, Gaussian noise, LD purification, and Adversarial attack. Additional sweeps analyze PSNR versus metric behavior, rank effects, step counts, and guidance scales, with runtimes summarized for key methods.",{"@graph":63,"@context":118},[64,80,101],{"@type":65,"itemListElement":66},"BreadcrumbList",[67,71,74,77],{"item":68,"name":69,"@type":70,"position":9},"https://docshare.wps.com","Home","ListItem",{"item":72,"name":10,"@type":70,"position":73},"https://docshare.wps.com/template/",2,{"item":75,"name":11,"@type":70,"position":76},"https://docshare.wps.com/template/presentations/",3,{"item":78,"name":59,"@type":70,"position":79},"https://docshare.wps.com/template/tfsr20250147-performance-metrics-table-for-aqualora-and-stable-diffusion/194473/",4,{"url":78,"name":59,"@type":81,"image":82,"author":87,"headline":59,"publisher":90,"fileFormat":93,"inLanguage":57,"description":61,"dateModified":94,"datePublished":95,"encodingFormat":93,"isAccessibleForFree":96,"interactionStatistic":97},"DigitalDocument",{"url":83,"@type":84,"width":85,"height":86},"https://docshare.wps.com/thumbnails/tfsr20250147-performance-metrics-table-for-aqualora-and-stable-diffusion/194473.png","ImageObject",442,249,{"name":88,"@type":89},"Seraphina","Person",{"url":68,"name":91,"@type":92},"DocShare","Organization","application/pdf","2026-09-20","2026-09-03",true,{"@type":98,"interactionType":99,"userInteractionCount":76},"InteractionCounter",{"@type":100},"ViewAction",{"@type":102,"mainEntity":103},"FAQPage",[104,110,114],{"name":105,"@type":106,"acceptedAnswer":107},"Which evaluation metrics are reported in the document tables?","Question",{"text":108,"@type":109},"The tables report ACC, TPR@1%FPR, PSNR, and SSIM for each method and setting.","Answer",{"name":111,"@type":106,"acceptedAnswer":112},"What perturbation or processing conditions are compared under AquaLoRA and Stable Diffusion?",{"text":113,"@type":109},"Comparisons include JPEG, Brightness, Gaussian noise, LD purification, and Adversarial attack, alongside an Original baseline.",{"name":115,"@type":106,"acceptedAnswer":116},"How does the document analyze the influence of diffusion settings?",{"text":117,"@type":109},"It provides separate tables varying PSNR-related thresholds, ranks, step counts, and guidance scales, each reporting the same set of metrics.","https://schema.org",{"og:url":78,"og:type":120,"og:title":59,"og:site_name":91,"og:description":61},"article",{"robots":122,"canonical":78},"index,follow",{"doc_id":124,"site_id":56},194473,1788439339,{"code":4,"msg":5,"data":127},{"doc_id":124,"user_id":128,"nickname":88,"user_avatar":129,"doc_module":9,"category_id":8,"category_name":11,"doc_title":59,"doc_description":61,"doc_content":130,"file_id":131,"file_url":132,"file_type":133,"file_size":134,"view_count":73,"is_deleted":4,"is_public":9,"is_downloadable":9,"audit_status":9,"page_count":135,"language":136,"language_code":57,"site_id":56,"html_lang":57,"table_of_contents":137,"faqs":138,"seo_title":139,"seo_description":61,"update_tm":125,"read_time":79},962075114101,"https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165","| Methods | AquaLoRA |  |  |  | Stable Signature |  |  |  | Tree-Ring |  |  |  |\n| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |\n|  | ACC | TPR@1%FPR | PSNR | SSIM | ACC | TPR@1%FPR | PSNR | SSIM | ACC | TPR@1%FPR | PSNR | SSIM |\n| Original | 0.862 | 0.998 | – | – | 0.998 | 1.000 | – | – | – | 1.000 | – | – |\n| JPEG | 0.842 | 0.992 | 33.868 | 0.903 | 0.814 | 0.998 | 30.651 | 0.851 | – | 0.012 | 16.358 | 0.424 |\n| Brightness | 0.660 | 0.504 | 4.592 | 0.210 | 0.921 | 0.960 | 12.932 | 0.706 | – | 0.418 | 16.192 | 0.410 |\n| Gaussian noise | 0.651 | 0.503 | 9.316 | 0.138 | 0.532 | 0.008 | 12.688 | 0.182 | – | 0.030 | 11.536 | 0.140 |\n| LD purification | 0.472 | 0.008 | 17.110 | 0.400 | 0.476 | 0.000 | 15.718 | 0.333 | – | 0.092 | 23.282 | 0.632 |\n| Adversarial attack | 0.785 | 0.970 | 33.593 | 0.923 | 0.998 | 1.000 | 38.183 | 0.975 | – | 0.000 | 16.013 | 0.537 |\n| Ours | 0.624 | 0.350 | 20.500 | 0.766 | 0.636 | 0.426 | 18.238 | 0.582 | – | 0.418 | 20.160 | 0.578 |\n\n\n| Methods | Runtime (s) |\n| --- | --- |\n| JPEG | 8 722.15 |\n| Brightness | 6 553.36 |\n| Gaussian noise | 6 963.45 |\n| LD purification | 10 737.13 |\n| Adversarial attack | 7 414.87 |\n| Ours | 6 078.24 |\n\n\n| Methods | Stable Diffusion v1.4 |  |  |  | Stable Diffusion v2.0 |  |  |  | Stable Diffusion v2.1 |  |  |  |\n| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |\n|  | ACC | TPR@1%FPR | PSNR | SSIM | ACC | TPR@1%FPR | PSNR | SSIM | ACC | TPR@1%FPR | PSNR | SSIM |\n| JPEG | 0.842 | 0.992 | 33.868 | 0.903 | 0.743 | 0.974 | 26.261 | 0.754 | 0.707 | 0.932 | 27.071 | 0.765 |\n| Brightness | 0.660 | 0.504 | 4.592 | 0.210 | 0.703 | 0.886 | 7.567 | 0.418 | 0.606 | 0.120 | 6.697 | 0.413 |\n| Gaussian noise | 0.651 | 0.503 | 9.316 | 0.138 | 0.578 | 0.000 | 8.383 | 0.200 | 0.534 | 0.000 | 7.882 | 0.170 |\n| LD purification | 0.472 | 0.008 | 17.110 | 0.400 | 0.635 | 0.309 | 12.432 | 0.321 | 0.578 | 0.024 | 12.319 | 0.413 |\n| Adversarial attack | 0.785 | 0.970 | 33.593 | 0.923 | 0.736 | 0.960 | 37.173 | 0.895 | 0.729 | 0.985 | 7.272 | 0.142 |\n| Ours | 0.624 | 0.350 | 20.500 | 0.766 | 0.665 | 0.575 | 18.238 | 0.582 | 0.647 | 0.352 | 15.802 | 0.750 |\n\n\n| PSNR | Metrics |  |  |  |\n| --- | --- | --- | --- | --- |\n|  | ACC | TPR@1%FPR | PSNR | SSIM |\n| Original | 0.862 | 0.998 | – | – |\n| 15 | 0.463 | 0.004 | 14.463 | 0.540 |\n| 20 | 0.631 | 0.373 | 21.227 | 0.743 |\n| 25 | 0.794 | 0.952 | 26.656 | 0.837 |\n| 30 | 0.813 | 0.981 | 31.547 | 0.958 |\n\n\n| Ranks | Metrics |  |  |  |\n| --- | --- | --- | --- | --- |\n|  | ACC | TPR@1%FPR | PSNR | SSIM |\n| Original | 0.862 | 0.998 | – | – |\n| 4 | 0.738 | 0.870 | 17.735 | 0.681 |\n| 16 | 0.624 | 0.350 | 20.500 | 0.766 |\n| 40 | 0.631 | 0.373 | 21.227 | 0.743 |\n| 80 | 0.601 | 0.244 | 20.435 | 0.670 |\n\n\n| Steps | Metrics |  |  |  |\n| --- | --- | --- | --- | --- |\n|  | ACC | TPR@1%FPR | PSNR | SSIM |\n| Original | 0.862 | 0.998 | – | – |\n| 15 | 0.621 | 0.351 | 19.368 | 0.667 |\n| 25 | 0.623 | 0.347 | 20.147 | 0.737 |\n| 50 | 0.624 | 0.350 | 20.500 | 0.766 |\n| 100 | 0.622 | 0.345 | 20.581 | 0.764 |\n\n\n| Guidance scales | Metrics |  |  |  |\n| --- | --- | --- | --- | --- |\n|  | ACC | TPR@1%FPR | PSNR | SSIM |\n| Original | 0.862 | 0.998 | – | – |\n| 5 | 0.604 | 0.254 | 19.323 | 0.653 |\n| 7.5 | 0.624 | 0.350 | 20.500 | 0.766 |\n| 10 | 0.639 | 0.430 | 19.058 | 0.631 |","cbCaisZJ5AL1kJKa","https://ap.wps.com/l/cbCaisZJ5AL1kJKa","pdf",1184738,10,"English","# Methods\n## AquaLoRA\n## Stable Diffusion variants\n## Metrics across PSNR, Ranks, Steps, and Guidance scales\n## Runtime comparison","[{\"question\":\"Which evaluation metrics are reported in the document tables?\",\"answer\":\"The tables report ACC, TPR@1%FPR, PSNR, and SSIM for each method and setting.\"},{\"question\":\"What perturbation or processing conditions are compared under AquaLoRA and Stable Diffusion?\",\"answer\":\"Comparisons include JPEG, Brightness, Gaussian noise, LD purification, and Adversarial attack, alongside an Original baseline.\"},{\"question\":\"How does the document analyze the influence of diffusion settings?\",\"answer\":\"It provides separate tables varying PSNR-related thresholds, ranks, step counts, and guidance scales, each reporting the same set of metrics.\"}]","tfsr20250147 - performance metrics table for AquaLoRA and Stable Diffusion | PDF"]