{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:H7VQ6FCC7VA4SMYSNSUNBXDXAF","short_pith_number":"pith:H7VQ6FCC","schema_version":"1.0","canonical_sha256":"3feb0f1442fd41c933126ca8d0dc7701466131d35aed7b3e8cc6adc7b88d50b8","source":{"kind":"arxiv","id":"2503.07699","version":2},"attestation_state":"computed","paper":{"title":"RayFlow: Instance-Aware Diffusion Acceleration via Adaptive Flow Trajectories","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Huiyang Shao, Xing Wang, Xin Xia, Xuefeng Xiao, Yuhong Yang, Yuxi Ren","submitted_at":"2025-03-10T17:20:52Z","abstract_excerpt":"Diffusion models have achieved remarkable success across various domains. However, their slow generation speed remains a critical challenge. Existing acceleration methods, while aiming to reduce steps, often compromise sample quality, controllability, or introduce training complexities. Therefore, we propose RayFlow, a novel diffusion framework that addresses these limitations. Unlike previous methods, RayFlow guides each sample along a unique path towards an instance-specific target distribution. This method minimizes sampling steps while preserving generation diversity and stability. Further"},"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":"2503.07699","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-03-10T17:20:52Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"2993911d45c0012810b3fd0c490fed82a2fce0ad3cb6d22c3600c512f07e5e1d","abstract_canon_sha256":"bd8d4e86524f807043cbcf5f408de8a6e64bfdbcba1d781f2b4212178412f57b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:38:44.104010Z","signature_b64":"HERU87tgGKC8wPhSGE+lACi5Ljszo0WY3cnoGtMus//NachfnTjWTjO5YwiKGMsdg8elAqjRUgNsdwgsknVHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3feb0f1442fd41c933126ca8d0dc7701466131d35aed7b3e8cc6adc7b88d50b8","last_reissued_at":"2026-07-05T10:38:44.103509Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:38:44.103509Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RayFlow: Instance-Aware Diffusion Acceleration via Adaptive Flow Trajectories","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Huiyang Shao, Xing Wang, Xin Xia, Xuefeng Xiao, Yuhong Yang, Yuxi Ren","submitted_at":"2025-03-10T17:20:52Z","abstract_excerpt":"Diffusion models have achieved remarkable success across various domains. However, their slow generation speed remains a critical challenge. Existing acceleration methods, while aiming to reduce steps, often compromise sample quality, controllability, or introduce training complexities. Therefore, we propose RayFlow, a novel diffusion framework that addresses these limitations. Unlike previous methods, RayFlow guides each sample along a unique path towards an instance-specific target distribution. This method minimizes sampling steps while preserving generation diversity and stability. Further"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.07699","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/2503.07699/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":"2503.07699","created_at":"2026-07-05T10:38:44.103566+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.07699v2","created_at":"2026-07-05T10:38:44.103566+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.07699","created_at":"2026-07-05T10:38:44.103566+00:00"},{"alias_kind":"pith_short_12","alias_value":"H7VQ6FCC7VA4","created_at":"2026-07-05T10:38:44.103566+00:00"},{"alias_kind":"pith_short_16","alias_value":"H7VQ6FCC7VA4SMYS","created_at":"2026-07-05T10:38:44.103566+00:00"},{"alias_kind":"pith_short_8","alias_value":"H7VQ6FCC","created_at":"2026-07-05T10:38:44.103566+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2512.13507","citing_title":"Seedance 1.5 pro: A Native Audio-Visual Joint Generation Foundation Model","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2504.11346","citing_title":"Seedream 3.0 Technical Report","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2509.20427","citing_title":"Seedream 4.0: Toward Next-generation Multimodal Image Generation","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2506.09113","citing_title":"Seedance 1.0: Exploring the Boundaries of Video Generation Models","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14910","citing_title":"Reward-Aware Trajectory Shaping for Few-step Visual Generation","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF","json":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF.json","graph_json":"https://pith.science/api/pith-number/H7VQ6FCC7VA4SMYSNSUNBXDXAF/graph.json","events_json":"https://pith.science/api/pith-number/H7VQ6FCC7VA4SMYSNSUNBXDXAF/events.json","paper":"https://pith.science/paper/H7VQ6FCC"},"agent_actions":{"view_html":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF","download_json":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF.json","view_paper":"https://pith.science/paper/H7VQ6FCC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.07699&json=true","fetch_graph":"https://pith.science/api/pith-number/H7VQ6FCC7VA4SMYSNSUNBXDXAF/graph.json","fetch_events":"https://pith.science/api/pith-number/H7VQ6FCC7VA4SMYSNSUNBXDXAF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF/action/storage_attestation","attest_author":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF/action/author_attestation","sign_citation":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF/action/citation_signature","submit_replication":"https://pith.science/pith/H7VQ6FCC7VA4SMYSNSUNBXDXAF/action/replication_record"}},"created_at":"2026-07-05T10:38:44.103566+00:00","updated_at":"2026-07-05T10:38:44.103566+00:00"}