{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/adversarial-score-identity-distillation","title":"Adversarial Score identity Distillation: Rapidly Surpassing the Teacher in One Step","arxiv_id":"2410.14919","date":"2024-10-19","proceeding":null,"authors":["Mingyuan Zhou","Huangjie Zheng","Yi Gu","Zhendong Wang","Hai Huang"],"abstract":"Score identity Distillation (SiD) is a data-free method that has achieved SOTA performance in image generation by leveraging only a pretrained diffusion model, without requiring any training data. However, its ultimate performance is constrained by how accurate the pretrained model captures the true data scores at different stages of the diffusion process. In this paper, we introduce SiDA (SiD with Adversarial Loss), which not only enhances generation quality but also improves distillation efficiency by incorporating real images and adversarial loss. SiDA utilizes the encoder from the generator's score network as a discriminator, allowing it to distinguish between real images and those generated by SiD. The adversarial loss is batch-normalized within each GPU and then combined with the original SiD loss. This integration effectively incorporates the average \"fakeness\" per GPU batch into the pixel-based SiD loss, enabling SiDA to distill a single-step generator. SiDA converges significantly faster than its predecessor when distilled from scratch, and swiftly improves upon the original model's performance during fine-tuning from a pre-distilled SiD generator. This one-step adversarial distillation method establishes new benchmarks in generation performance when distilling EDM diffusion models, achieving FID scores of 1.110 on ImageNet 64x64. When distilling EDM2 models trained on ImageNet 512x512, our SiDA method surpasses even the largest teacher model, EDM2-XXL, which achieved an FID of 1.81 using classifier-free guidance (CFG) and 63 generation steps. In contrast, SiDA achieves FID scores of 2.156 for size XS, 1.669 for S, 1.488 for M, 1.413 for L, 1.379 for XL, and 1.366 for XXL, all without CFG and in a single generation step. These results highlight substantial improvements across all model sizes. Our code is available at https://github.com/mingyuanzhou/SiD/tree/sida.","url_abs":"https://arxiv.org/abs/2410.14919v4","url_pdf":"https://arxiv.org/pdf/2410.14919v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"adversarial-score-identity-distillation","repo_url":"https://github.com/mingyuanzhou/sid","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"adversarial-score-identity-distillation","repo_url":"https://github.com/mingyuanzhou/sid-lsg","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"conditional-image-generation","task_name":"Conditional Image Generation"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"unconditional-image-generation","task_name":"Unconditional Image Generation"}],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-generation-on-afhq-v2-64x64","task":"Image Generation","dataset":"AFHQ-v2 64x64","model":"SiDA-EDM","rank_in_archive_order":1,"of":2,"metrics":{"FID":"1.276","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-cifar-10","task":"Image Generation","dataset":"CIFAR-10","model":"SiDA-EDM","rank_in_archive_order":5,"of":78,"metrics":{"FID":"1.396","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-ffhq-64x64","task":"Image Generation","dataset":"FFHQ 64x64","model":"SiDA-EDM","rank_in_archive_order":1,"of":2,"metrics":{"FID":"1.040","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiDA-EDM2-XXL (1.5B)","rank_in_archive_order":7,"of":52,"metrics":{"FID":"1.366","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiDA-EDM2-XL (1.1B)","rank_in_archive_order":8,"of":52,"metrics":{"FID":"1.379","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiDA-EDM2-L (777M)","rank_in_archive_order":10,"of":52,"metrics":{"FID":"1.413","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiDA-EDM2-M (498M)","rank_in_archive_order":12,"of":52,"metrics":{"FID":"1.488","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiDA-EDM2-S (280M)","rank_in_archive_order":13,"of":52,"metrics":{"FID":"1.669","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiD-EDM2-XL (1.1B)","rank_in_archive_order":23,"of":52,"metrics":{"FID":"1.888","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiD-EDM2-L (777M)","rank_in_archive_order":24,"of":52,"metrics":{"FID":"1.907","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiD-EDM2-XXL (1.5B)","rank_in_archive_order":26,"of":52,"metrics":{"FID":"1.969","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiD-EDM2-M (498M)","rank_in_archive_order":28,"of":52,"metrics":{"FID":"2.06","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiDA-EDM2-XS (125M)","rank_in_archive_order":30,"of":52,"metrics":{"FID":"2.156","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiD-EDM2-S (280M)","rank_in_archive_order":36,"of":52,"metrics":{"FID":"2.707","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-512x512","task":"Image Generation","dataset":"ImageNet 512x512","model":"SiD-EDM2-XS (125M)","rank_in_archive_order":43,"of":52,"metrics":{"FID":"3.353","NFE":"1"},"uses_additional_data":false},{"leaderboard":"/sota/image-generation-on-imagenet-64x64","task":"Image Generation","dataset":"ImageNet 64x64","model":"SiDA-EDM","rank_in_archive_order":4,"of":65,"metrics":{"FID":"1.11","NFE":"1"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2410.14919","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.14919"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mingyuanzhou/sid-lsg","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mingyuanzhou/sid","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_honours":2,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":1,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"68b5e143931956cf","entry":"calculate_fid_from_inception_stats","repo":"mingyuanzhou/sid","repo_kind":"official","path":"sid_generate.py","file_url":"https://github.com/mingyuanzhou/sid/blob/HEAD/sid_generate.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"68b5e143931956cf"}},{"code_sha256_prefix":"cacd4f6ec202d9b4","entry":"parse_int_list","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"cacd4f6ec202d9b4"}},{"code_sha256_prefix":"4c9c2d2ca6c238d8","entry":"sid_sampler","repo":"mingyuanzhou/sid","repo_kind":"official","path":"sid_generate.py","file_url":"https://github.com/mingyuanzhou/sid/blob/HEAD/sid_generate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4c9c2d2ca6c238d8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}