{"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/a-closer-look-at-time-steps-is-worthy-of","title":"A Closer Look at Time Steps is Worthy of Triple Speed-Up for Diffusion Model Training","arxiv_id":"2405.17403","date":"2024-05-27","proceeding":"CVPR 2025 1","authors":["Kai Wang","Mingjia Shi","Yukun Zhou","Zekai Li","Zhihang Yuan","Yuzhang Shang","Xiaojiang Peng","Hanwang Zhang","Yang You"],"abstract":"Training diffusion models is always a computation-intensive task. In this paper, we introduce a novel speed-up method for diffusion model training, called, which is based on a closer look at time steps. Our key findings are: i) Time steps can be empirically divided into acceleration, deceleration, and convergence areas based on the process increment. ii) These time steps are imbalanced, with many concentrated in the convergence area. iii) The concentrated steps provide limited benefits for diffusion training. To address this, we design an asymmetric sampling strategy that reduces the frequency of steps from the convergence area while increasing the sampling probability for steps from other areas. Additionally, we propose a weighting strategy to emphasize the importance of time steps with rapid-change process increments. As a plug-and-play and architecture-agnostic approach, SpeeD consistently achieves 3-times acceleration across various diffusion architectures, datasets, and tasks. Notably, due to its simple design, our approach significantly reduces the cost of diffusion model training with minimal overhead. Our research enables more researchers to train diffusion models at a lower cost.","url_abs":"https://arxiv.org/abs/2405.17403v3","url_pdf":"https://arxiv.org/pdf/2405.17403v3.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":"a-closer-look-at-time-steps-is-worthy-of","repo_url":"https://github.com/1zeryu/speed","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"a-closer-look-at-time-steps-is-worthy-of","repo_url":"https://github.com/nus-hpc-ai-lab/speed","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[{"method_slug":"diffusion","method_name":"Diffusion"},{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2405.17403","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.17403"}},"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/1zeryu/speed","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/nus-hpc-ai-lab/speed","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_draft_wrong":2,"ran_fixture":1,"ran_honours":2,"ran":1,"unverified":5},"by_repo_kind":{"official":{"samples":11,"ran":6,"repositories":2}},"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":0,"samples":[{"code_sha256_prefix":"1712a07966b542ee","entry":"center_crop_arr","repo":"1zeryu/speed","repo_kind":"official","path":"speed/dataset/transform.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/dataset/transform.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1712a07966b542ee"}},{"code_sha256_prefix":"e5947aba1d10885f","entry":"get_1d_sincos_pos_embed_from_grid","repo":"1zeryu/speed","repo_kind":"official","path":"speed/networks/dit/mdt.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/networks/dit/mdt.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e5947aba1d10885f"}},{"code_sha256_prefix":"c92c27c924b517e8","entry":"get_2d_sincos_pos_embed","repo":"1zeryu/speed","repo_kind":"official","path":"speed/networks/dit/mdt.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/networks/dit/mdt.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c92c27c924b517e8"}},{"code_sha256_prefix":"debdeb32aa1aa3e0","entry":"get_image_transform","repo":"1zeryu/speed","repo_kind":"official","path":"speed/dataset/transform.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/dataset/transform.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"debdeb32aa1aa3e0"}},{"code_sha256_prefix":"03310bba324ae4fb","entry":"modulate","repo":"1zeryu/speed","repo_kind":"official","path":"speed/networks/dit/dit.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/networks/dit/dit.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"03310bba324ae4fb"}},{"code_sha256_prefix":"a5e3d9618aac5205","entry":"t2i_modulate","repo":"nus-hpc-ai-lab/speed","repo_kind":"official","path":"speed/networks/pixart/PixArt_blocks.py","file_url":"https://github.com/nus-hpc-ai-lab/speed/blob/HEAD/speed/networks/pixart/PixArt_blocks.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a5e3d9618aac5205"}},{"code_sha256_prefix":"7700f3428f58afbe","entry":"get_1d_sincos_pos_embed_from_grid","repo":"1zeryu/speed","repo_kind":"official","path":"speed/networks/pixart/PixArt.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/networks/pixart/PixArt.py","link_basis":"harvester_set","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":"7700f3428f58afbe"}},{"code_sha256_prefix":"5a10b94b2fe962d2","entry":"get_2d_sincos_pos_embed","repo":"1zeryu/speed","repo_kind":"official","path":"speed/networks/pixart/PixArt.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/networks/pixart/PixArt.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":"5a10b94b2fe962d2"}},{"code_sha256_prefix":"60599713933af278","entry":"get_2d_sincos_pos_embed_from_grid","repo":"nus-hpc-ai-lab/speed","repo_kind":"official","path":"speed/networks/pixart/PixArt.py","file_url":"https://github.com/nus-hpc-ai-lab/speed/blob/HEAD/speed/networks/pixart/PixArt.py","link_basis":"plan_row","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":"60599713933af278"}},{"code_sha256_prefix":"665d8a4e8f673a4c","entry":"get_2d_sincos_pos_embed_from_grid","repo":"1zeryu/speed","repo_kind":"official","path":"speed/networks/dit/mdt.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/networks/dit/mdt.py","link_basis":"harvester_set","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":"665d8a4e8f673a4c"}},{"code_sha256_prefix":"8849fb1444c6af62","entry":"image_dataset","repo":"1zeryu/speed","repo_kind":"official","path":"speed/dataset/image.py","file_url":"https://github.com/1zeryu/speed/blob/HEAD/speed/dataset/image.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":"8849fb1444c6af62"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}