{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3MCGYVRFISTVIYDSOH2GIHAGTU","short_pith_number":"pith:3MCGYVRF","schema_version":"1.0","canonical_sha256":"db046c562544a754607271f4641c069d28d1b0166942636dd294a6439936901d","source":{"kind":"arxiv","id":"2406.14548","version":2},"attestation_state":"computed","paper":{"title":"Consistency Models Made Easy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ashwini Pokle, Justin Lin, J. Zico Kolter, William Luo, Zhengyang Geng","submitted_at":"2024-06-20T17:56:02Z","abstract_excerpt":"Consistency models (CMs) offer faster sampling than traditional diffusion models, but their training is resource-intensive. For example, as of 2024, training a state-of-the-art CM on CIFAR-10 takes one week on 8 GPUs. In this work, we propose an effective scheme for training CMs that largely improves the efficiency of building such models. Specifically, by expressing CM trajectories via a particular differential equation, we argue that diffusion models can be viewed as a special case of CMs. We can thus fine-tune a consistency model starting from a pretrained diffusion model and progressively "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2406.14548","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-20T17:56:02Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"c62e7987d1da27e7a816234b190de2b5dfba1809ac4080f13436b3c7792fa314","abstract_canon_sha256":"4a00976b1572e07c3c14731804caceff62f021022f468f413c8d72f536034f7e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:52.909246Z","signature_b64":"H0/RpbLWSX7BCkTF6WvN+zaLVWKBeTKkGROMfUluuUWrjIcGIS9Ge3w12SI2NSUgs6XGOU2egsVCLkFE7Xd9AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"db046c562544a754607271f4641c069d28d1b0166942636dd294a6439936901d","last_reissued_at":"2026-07-05T09:18:52.908775Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:52.908775Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Consistency Models Made Easy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ashwini Pokle, Justin Lin, J. Zico Kolter, William Luo, Zhengyang Geng","submitted_at":"2024-06-20T17:56:02Z","abstract_excerpt":"Consistency models (CMs) offer faster sampling than traditional diffusion models, but their training is resource-intensive. For example, as of 2024, training a state-of-the-art CM on CIFAR-10 takes one week on 8 GPUs. In this work, we propose an effective scheme for training CMs that largely improves the efficiency of building such models. Specifically, by expressing CM trajectories via a particular differential equation, we argue that diffusion models can be viewed as a special case of CMs. We can thus fine-tune a consistency model starting from a pretrained diffusion model and progressively "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.14548","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2406.14548/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2406.14548","created_at":"2026-07-05T09:18:52.908831+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.14548v2","created_at":"2026-07-05T09:18:52.908831+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.14548","created_at":"2026-07-05T09:18:52.908831+00:00"},{"alias_kind":"pith_short_12","alias_value":"3MCGYVRFISTV","created_at":"2026-07-05T09:18:52.908831+00:00"},{"alias_kind":"pith_short_16","alias_value":"3MCGYVRFISTVIYDS","created_at":"2026-07-05T09:18:52.908831+00:00"},{"alias_kind":"pith_short_8","alias_value":"3MCGYVRF","created_at":"2026-07-05T09:18:52.908831+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":21,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00535","citing_title":"Flow-Map GRPO: Reinforcement Learning for Few-Step Flow-Map Generators via Anchored Stochastic Composition","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05254","citing_title":"Flash-WAM: Modality-Aware Distillation for World Action Models","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27147","citing_title":"How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02214","citing_title":"Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2511.11688","citing_title":"Hierarchical Schedule Optimization for Fast and Robust Diffusion Model Sampling","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2602.02214","citing_title":"Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27147","citing_title":"How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17899","citing_title":"DCFold: Efficient Protein Structure Generation with Single Forward Pass","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2510.08431","citing_title":"Large Scale Diffusion Distillation via Score-Regularized Continuous-Time Consistency","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2512.23709","citing_title":"Stream-DiffVSR: Low-Latency Streamable Video Super-Resolution via Auto-Regressive Diffusion","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2602.10764","citing_title":"Dual-End Consistency Model","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2410.12557","citing_title":"One Step Diffusion via Shortcut Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13724","citing_title":"AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27147","citing_title":"How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08115","citing_title":"Alice v1: Distillation-Enhanced Video Generation Surpassing Closed-Source Models","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03623","citing_title":"A Few-Step Generative Model on Cumulative Flow Maps","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2505.13447","citing_title":"Mean Flows for One-step Generative Modeling","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12273","citing_title":"SubFlow: Sub-mode Conditioned Flow Matching for Diverse One-Step Generation","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08960","citing_title":"Efficient Hierarchical Implicit Flow Q-learning for Offline Goal-conditioned Reinforcement Learning","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08837","citing_title":"Discrete Meanflow Training Curriculum","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18168","citing_title":"Extending One-Step Image Generation from Class Labels to Text via Discriminative Text Representation","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU","json":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU.json","graph_json":"https://pith.science/api/pith-number/3MCGYVRFISTVIYDSOH2GIHAGTU/graph.json","events_json":"https://pith.science/api/pith-number/3MCGYVRFISTVIYDSOH2GIHAGTU/events.json","paper":"https://pith.science/paper/3MCGYVRF"},"agent_actions":{"view_html":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU","download_json":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU.json","view_paper":"https://pith.science/paper/3MCGYVRF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.14548&json=true","fetch_graph":"https://pith.science/api/pith-number/3MCGYVRFISTVIYDSOH2GIHAGTU/graph.json","fetch_events":"https://pith.science/api/pith-number/3MCGYVRFISTVIYDSOH2GIHAGTU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU/action/storage_attestation","attest_author":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU/action/author_attestation","sign_citation":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU/action/citation_signature","submit_replication":"https://pith.science/pith/3MCGYVRFISTVIYDSOH2GIHAGTU/action/replication_record"}},"created_at":"2026-07-05T09:18:52.908831+00:00","updated_at":"2026-07-05T09:18:52.908831+00:00"}