{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:2QPDSNZGFPSKX6UNBAYGNOERV7","short_pith_number":"pith:2QPDSNZG","schema_version":"1.0","canonical_sha256":"d41e3937262be4abfa8d083066b891aff2b8d3cfe1cf01e4558a40879fde758d","source":{"kind":"arxiv","id":"2504.19353","version":1},"attestation_state":"computed","paper":{"title":"Flow Along the K-Amplitude for Generative Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fan Wang, Jiasheng Tang, Shengchao Liu, Shuning Chang, Weitao Du, Yu Rong","submitted_at":"2025-04-27T20:38:24Z","abstract_excerpt":"In this work, we propose a novel generative learning paradigm, K-Flow, an algorithm that flows along the $K$-amplitude. Here, $k$ is a scaling parameter that organizes frequency bands (or projected coefficients), and amplitude describes the norm of such projected coefficients. By incorporating the $K$-amplitude decomposition, K-Flow enables flow matching across the scaling parameter as time. We discuss three venues and six properties of K-Flow, from theoretical foundations, energy and temporal dynamics, and practical applications, respectively. Specifically, from the practical usage perspectiv"},"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":"2504.19353","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-04-27T20:38:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"52a0d1ce4d0389a04b81571116219996e3bb80feab74d54369786006c3a8d97c","abstract_canon_sha256":"a46778afc3b8c0ccdefac7ea0d350109297c2caaddc6cf74898efaf2ce13ca4a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:54:56.106202Z","signature_b64":"CgQ/hPeVnX9OzuSU7VFA0YxwwFqDZEbXDExEeUDOw8PPrUDNUuBYb7/uryBhVwJR34Pxj7V87qeTe2iB8LogCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d41e3937262be4abfa8d083066b891aff2b8d3cfe1cf01e4558a40879fde758d","last_reissued_at":"2026-07-05T10:54:56.105708Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:54:56.105708Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Flow Along the K-Amplitude for Generative Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fan Wang, Jiasheng Tang, Shengchao Liu, Shuning Chang, Weitao Du, Yu Rong","submitted_at":"2025-04-27T20:38:24Z","abstract_excerpt":"In this work, we propose a novel generative learning paradigm, K-Flow, an algorithm that flows along the $K$-amplitude. Here, $k$ is a scaling parameter that organizes frequency bands (or projected coefficients), and amplitude describes the norm of such projected coefficients. By incorporating the $K$-amplitude decomposition, K-Flow enables flow matching across the scaling parameter as time. We discuss three venues and six properties of K-Flow, from theoretical foundations, energy and temporal dynamics, and practical applications, respectively. Specifically, from the practical usage perspectiv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.19353","kind":"arxiv","version":1},"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/2504.19353/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":"2504.19353","created_at":"2026-07-05T10:54:56.105768+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.19353v1","created_at":"2026-07-05T10:54:56.105768+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.19353","created_at":"2026-07-05T10:54:56.105768+00:00"},{"alias_kind":"pith_short_12","alias_value":"2QPDSNZGFPSK","created_at":"2026-07-05T10:54:56.105768+00:00"},{"alias_kind":"pith_short_16","alias_value":"2QPDSNZGFPSKX6UN","created_at":"2026-07-05T10:54:56.105768+00:00"},{"alias_kind":"pith_short_8","alias_value":"2QPDSNZG","created_at":"2026-07-05T10:54:56.105768+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.18736","citing_title":"Spectral Progressive Diffusion for Efficient Image and Video Generation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18736","citing_title":"Spectral Progressive Diffusion for Efficient Image and Video Generation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20902","citing_title":"Frequency-Forcing: From Scaling-as-Time to Soft Frequency Guidance","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09979","citing_title":"A Minimal Model of Representation Collapse: Frustration, Stop-Gradient, and Dynamics","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7","json":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7.json","graph_json":"https://pith.science/api/pith-number/2QPDSNZGFPSKX6UNBAYGNOERV7/graph.json","events_json":"https://pith.science/api/pith-number/2QPDSNZGFPSKX6UNBAYGNOERV7/events.json","paper":"https://pith.science/paper/2QPDSNZG"},"agent_actions":{"view_html":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7","download_json":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7.json","view_paper":"https://pith.science/paper/2QPDSNZG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.19353&json=true","fetch_graph":"https://pith.science/api/pith-number/2QPDSNZGFPSKX6UNBAYGNOERV7/graph.json","fetch_events":"https://pith.science/api/pith-number/2QPDSNZGFPSKX6UNBAYGNOERV7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7/action/storage_attestation","attest_author":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7/action/author_attestation","sign_citation":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7/action/citation_signature","submit_replication":"https://pith.science/pith/2QPDSNZGFPSKX6UNBAYGNOERV7/action/replication_record"}},"created_at":"2026-07-05T10:54:56.105768+00:00","updated_at":"2026-07-05T10:54:56.105768+00:00"}