{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Z2WTNZKJMXWEIQTFH4GJ2XATUW","short_pith_number":"pith:Z2WTNZKJ","schema_version":"1.0","canonical_sha256":"cead36e54965ec4442653f0c9d5c13a59ac8aa4ee3297bbc3042e24346e85057","source":{"kind":"arxiv","id":"2602.07928","version":2},"attestation_state":"computed","paper":{"title":"A Kinetic Energy Perspective of Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Enmao Diao, Fei Gao, Henrik Bostrom, Huancheng Hu, Michalis Vazirgiannis, Soon Hoe Lim, Xuyu Li, Zezhen Ding, Ziyun Li","submitted_at":"2026-02-08T11:51:50Z","abstract_excerpt":"Flow-based generative models can be viewed through a physics lens: sampling transports a particle from noise to data by integrating a learned velocity field, and each sample corresponds to a trajectory with its own dynamical effort. Motivated by classical mechanics, we introduce Kinetic Path Energy (KPE), an action-like, per-sample diagnostic that measures the accumulated kinetic effort along an ordinary differential equation (ODE) trajectory. Empirically, KPE exhibits two robust correspondences: {i} higher KPE predicts stronger semantic fidelity; {ii} high-KPE trajectories land in sparse repr"},"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":"2602.07928","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-08T11:51:50Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9eef4761d499bdd4ca3d5a512ff508bf9f9f8ae86a2a7282d1d1dd31095531cf","abstract_canon_sha256":"2f15211aa5402a568d4cae66b9caa0e45056ba8b53b43a09605fce9438aa8b86"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:32.977540Z","signature_b64":"TsLIaStZToznJOAh25GgeM+dieucKrrb/qPhhTagVO+gvHTkbjl0MTD7SEIRPF5vSWoOuJFNOeyRaJzVTPmfAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cead36e54965ec4442653f0c9d5c13a59ac8aa4ee3297bbc3042e24346e85057","last_reissued_at":"2026-06-01T01:02:32.976423Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:32.976423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Kinetic Energy Perspective of Flow Matching","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Enmao Diao, Fei Gao, Henrik Bostrom, Huancheng Hu, Michalis Vazirgiannis, Soon Hoe Lim, Xuyu Li, Zezhen Ding, Ziyun Li","submitted_at":"2026-02-08T11:51:50Z","abstract_excerpt":"Flow-based generative models can be viewed through a physics lens: sampling transports a particle from noise to data by integrating a learned velocity field, and each sample corresponds to a trajectory with its own dynamical effort. Motivated by classical mechanics, we introduce Kinetic Path Energy (KPE), an action-like, per-sample diagnostic that measures the accumulated kinetic effort along an ordinary differential equation (ODE) trajectory. Empirically, KPE exhibits two robust correspondences: {i} higher KPE predicts stronger semantic fidelity; {ii} high-KPE trajectories land in sparse repr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.07928","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/2602.07928/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":"2602.07928","created_at":"2026-06-01T01:02:32.976581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.07928v2","created_at":"2026-06-01T01:02:32.976581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.07928","created_at":"2026-06-01T01:02:32.976581+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z2WTNZKJMXWE","created_at":"2026-06-01T01:02:32.976581+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z2WTNZKJMXWEIQTF","created_at":"2026-06-01T01:02:32.976581+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z2WTNZKJ","created_at":"2026-06-01T01:02:32.976581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.11547","citing_title":"Sharpen Your Flow: Sharpness-Aware Sampling for Flow Matching","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW","json":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW.json","graph_json":"https://pith.science/api/pith-number/Z2WTNZKJMXWEIQTFH4GJ2XATUW/graph.json","events_json":"https://pith.science/api/pith-number/Z2WTNZKJMXWEIQTFH4GJ2XATUW/events.json","paper":"https://pith.science/paper/Z2WTNZKJ"},"agent_actions":{"view_html":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW","download_json":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW.json","view_paper":"https://pith.science/paper/Z2WTNZKJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.07928&json=true","fetch_graph":"https://pith.science/api/pith-number/Z2WTNZKJMXWEIQTFH4GJ2XATUW/graph.json","fetch_events":"https://pith.science/api/pith-number/Z2WTNZKJMXWEIQTFH4GJ2XATUW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW/action/storage_attestation","attest_author":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW/action/author_attestation","sign_citation":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW/action/citation_signature","submit_replication":"https://pith.science/pith/Z2WTNZKJMXWEIQTFH4GJ2XATUW/action/replication_record"}},"created_at":"2026-06-01T01:02:32.976581+00:00","updated_at":"2026-06-01T01:02:32.976581+00:00"}