{"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-2509-24526","title":"CMT: Mid-Training for Efficient Learning of Consistency, Mean Flow, and Flow Map Models","arxiv_id":"2509.24526","date":"2025-09-29","proceeding":"ICLR","authors":["Zheyuan Hu","Chieh-Hsin Lai","Yuki Mitsufuji","Stefano Ermon"],"abstract":"Flow map models such as Consistency Models (CM) and Mean Flow (MF) enable few-step generation by learning the long jump of the ODE solution of diffusion models, yet training remains unstable, sensitive to hyperparameters, and costly. Initializing from a pre-trained diffusion model helps, but still requires converting infinitesimal steps into a long-jump map, leaving instability unresolved. We introduce mid-training, the first concept and practical method that inserts a lightweight intermediate stage between the (diffusion) pre-training and the final flow map training (i.e., post-training) for vision generation. Concretely, Consistency Mid-Training (CMT) is a compact and principled stage that trains a model to map points along a solver trajectory from a pre-trained model, starting from a prior sample, directly to the solver-generated clean sample. It yields a trajectory-consistent and stable initialization. This initializer outperforms random and diffusion-based baselines and enables fast, robust convergence without heuristics. Initializing post-training with CMT weights further simplifies flow map learning. Empirically, CMT achieves state of the art two step FIDs: 1.97 on CIFAR-10, 1.32 on ImageNet 64x64, and 1.84 on ImageNet 512x512, while using up to 98% less training data and GPU time, compared to CMs. On ImageNet 256x256, CMT reaches 1-step FID 3.34 while cutting total training time by about 50% compared to MF from scratch (FID 3.43). This establishes CMT as a principled, efficient, and general framework for training flow map models.","url_abs":"https://arxiv.org/abs/2509.24526","url_pdf":"https://arxiv.org/pdf/2509.24526","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=2509.24526","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2509.24526"}},"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/sony/cmt","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":3,"ran_honours":1,"ran_fixture":1,"unverified":5},"by_repo_kind":{"found_in_text":{"samples":10,"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":"674eca7b9e1b6439","entry":"format_time","repo":"sony/cmt","repo_kind":"found_in_text","path":"dnnlib/util.py","file_url":"https://github.com/sony/cmt/blob/HEAD/dnnlib/util.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"674eca7b9e1b6439"}},{"code_sha256_prefix":"32af001972a0699f","entry":"format_time_brief","repo":"sony/cmt","repo_kind":"found_in_text","path":"dnnlib/util.py","file_url":"https://github.com/sony/cmt/blob/HEAD/dnnlib/util.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"32af001972a0699f"}},{"code_sha256_prefix":"cacd4f6ec202d9b4","entry":"parse_int_list","repo":"sony/cmt","repo_kind":"found_in_text","path":"run_cmt.py","file_url":"https://github.com/sony/cmt/blob/HEAD/run_cmt.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cacd4f6ec202d9b4"}},{"code_sha256_prefix":"a1023a0d09eca659","entry":"setup_snapshot_image_grid","repo":"sony/cmt","repo_kind":"found_in_text","path":"training/cmt_training_loop.py","file_url":"https://github.com/sony/cmt/blob/HEAD/training/cmt_training_loop.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a1023a0d09eca659"}},{"code_sha256_prefix":"d41a4250066bce93","entry":"weight_init","repo":"sony/cmt","repo_kind":"found_in_text","path":"training/networks.py","file_url":"https://github.com/sony/cmt/blob/HEAD/training/networks.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"d41a4250066bce93"}},{"code_sha256_prefix":"9d31d2c4cd16bb2d","entry":"ask_yes_no","repo":"sony/cmt","repo_kind":"found_in_text","path":"dnnlib/util.py","file_url":"https://github.com/sony/cmt/blob/HEAD/dnnlib/util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9d31d2c4cd16bb2d"}},{"code_sha256_prefix":"47c7a1249e324f74","entry":"compute_is","repo":"sony/cmt","repo_kind":"found_in_text","path":"metrics/inception_score.py","file_url":"https://github.com/sony/cmt/blob/HEAD/metrics/inception_score.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"47c7a1249e324f74"}},{"code_sha256_prefix":"ccfabb29e5375ee5","entry":"generator_fn","repo":"sony/cmt","repo_kind":"found_in_text","path":"training/cmt_training_loop.py","file_url":"https://github.com/sony/cmt/blob/HEAD/training/cmt_training_loop.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ccfabb29e5375ee5"}},{"code_sha256_prefix":"c127dc46d6966547","entry":"get_feature_detector_name","repo":"sony/cmt","repo_kind":"found_in_text","path":"metrics/metric_utils.py","file_url":"https://github.com/sony/cmt/blob/HEAD/metrics/metric_utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c127dc46d6966547"}},{"code_sha256_prefix":"67f3d7aa67fc48b6","entry":"learning_rate_schedule","repo":"sony/cmt","repo_kind":"found_in_text","path":"training/cmt_training_loop.py","file_url":"https://github.com/sony/cmt/blob/HEAD/training/cmt_training_loop.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"67f3d7aa67fc48b6"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CV","source":"arxiv_api"},"syntology_extracted_results":null}