{"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/arxiv-2605-30116","title":"SGMD: Score Gradient Matching Distillation for Few-Step Video Diffusion Distillation","arxiv_id":"2605.30116","date":"2026-05-28","proceeding":"ICML","authors":["Zhuguanyu Wu","Ruihao Gong","Yang Yong","Yushi Huang","Xiangyu Fan","Lei Yang","Dahua Lin","Xianglong Liu"],"abstract":"Distribution Matching Distillation (DMD) is a widely used paradigm for accelerating inference in few-step video diffusion models. However, DMD-style video distillation faces two coupled challenges: the fake score must track a continuously evolving generator, making training costly when frequent updates are required, while reverse-KL-style matching can be mode-seeking and conservative for preserving strong motion dynamics. To address these issues, we propose \\textbf{Score Gradient Matching Distillation (SGMD)}. SGMD adopts a fake-score perspective by directly optimizing the fake score toward the teacher, while using teacher stop-gradient Fisher as a stable distribution-matching objective. We provide a gradient analysis that motivates this objective choice under ideal tracking. Building on this, SGMD introduces a pair of dual potentials: negative-residual (NR) for outer-loop correction and residual-contraction (RC) for inner-loop tracking. Empirically, compared to DMD2, SGMD achieves an approximately $\\sim 3\\times$ training speedup and substantially improves motion dynamics for 4-step distilled models while preserving temporal consistency. A human study confirms that SGMD is preferred in motion quality and overall preference, while visual quality and text alignment remain comparable. Code is available at https://github.com/ModelTC/LightX2V.","url_abs":"https://arxiv.org/abs/2605.30116","url_pdf":"https://arxiv.org/pdf/2605.30116","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2605.30116","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2605.30116"}},"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":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/ModelTC/LightX2V","reach":null}],"summary":{"ran":5,"unverified":1},"by_repo_kind":{"found_in_text":{"samples":6,"ran":5,"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":0,"samples":[{"code_sha256_prefix":"7615179034d1e782","entry":"dynamic_metadata","repo":"ModelTC/LightX2V","repo_kind":"found_in_text","path":"lightx2v_kernel_xpu/build_version.py","file_url":"https://github.com/ModelTC/LightX2V/blob/HEAD/lightx2v_kernel_xpu/build_version.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":"7615179034d1e782"}},{"code_sha256_prefix":"0cbcc1bd5c7993ca","entry":"fi_force_retune","repo":"ModelTC/LightX2V","repo_kind":"found_in_text","path":"lightx2v/common/flashinfer_autotune.py","file_url":"https://github.com/ModelTC/LightX2V/blob/HEAD/lightx2v/common/flashinfer_autotune.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":"0cbcc1bd5c7993ca"}},{"code_sha256_prefix":"6194816c4482a0fe","entry":"get_qkv_subset","repo":"ModelTC/LightX2V","repo_kind":"found_in_text","path":"lightx2v/utils/print_atten_score.py","file_url":"https://github.com/ModelTC/LightX2V/blob/HEAD/lightx2v/utils/print_atten_score.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":"6194816c4482a0fe"}},{"code_sha256_prefix":"750c36f042d67fdb","entry":"group_concurrent_contiguous","repo":"ModelTC/LightX2V","repo_kind":"found_in_text","path":"lightx2v/disagg/conn.py","file_url":"https://github.com/ModelTC/LightX2V/blob/HEAD/lightx2v/disagg/conn.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":"750c36f042d67fdb"}},{"code_sha256_prefix":"68fcdfd881d93961","entry":"scaled_dot_product_attention","repo":"ModelTC/LightX2V","repo_kind":"found_in_text","path":"lightx2v/utils/print_atten_score.py","file_url":"https://github.com/ModelTC/LightX2V/blob/HEAD/lightx2v/utils/print_atten_score.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":"68fcdfd881d93961"}},{"code_sha256_prefix":"187b6079a8eddec8","entry":"fi_autotune_cache_path","repo":"ModelTC/LightX2V","repo_kind":"found_in_text","path":"lightx2v/common/flashinfer_autotune.py","file_url":"https://github.com/ModelTC/LightX2V/blob/HEAD/lightx2v/common/flashinfer_autotune.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":"187b6079a8eddec8"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.LG","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}