{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DVU6JAHPHJQFLYK6T73QSXPYWH","short_pith_number":"pith:DVU6JAHP","schema_version":"1.0","canonical_sha256":"1d69e480ef3a6055e15e9ff7095df8b1e7aca9a77b31b316b37390de0b21aab0","source":{"kind":"arxiv","id":"2412.03487","version":1},"attestation_state":"computed","paper":{"title":"Flow Matching with General Discrete Paths: A Kinetic-Optimal Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anuroop Sriram, Brian Karrer, Daniel Severo, Itai Gat, Marton Havasi, Neta Shaul, Peter Holderrieth, Ricky T. Q. Chen, Yaron Lipman","submitted_at":"2024-12-04T17:24:35Z","abstract_excerpt":"The design space of discrete-space diffusion or flow generative models are significantly less well-understood than their continuous-space counterparts, with many works focusing only on a simple masked construction. In this work, we aim to take a holistic approach to the construction of discrete generative models based on continuous-time Markov chains, and for the first time, allow the use of arbitrary discrete probability paths, or colloquially, corruption processes. Through the lens of optimizing the symmetric kinetic energy, we propose velocity formulas that can be applied to any given proba"},"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":"2412.03487","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-04T17:24:35Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b5f2860818a504b1edcc2875841eeb3ada19a6e5201b99040f26c954095bf1fd","abstract_canon_sha256":"4dd0da732648424f498ba11dfa006a4454cc084cbb09509a0c520c70cf1906c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:44:30.189224Z","signature_b64":"eU0plxooOLO26HsZ9K3nASxhwC+sESMLULkKkhehyhYjvKWl5CoEIR1ZdxAu3kVcTpVfM8Ma4YHDuv9bwsr8DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1d69e480ef3a6055e15e9ff7095df8b1e7aca9a77b31b316b37390de0b21aab0","last_reissued_at":"2026-07-05T09:44:30.188644Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:44:30.188644Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Flow Matching with General Discrete Paths: A Kinetic-Optimal Perspective","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Anuroop Sriram, Brian Karrer, Daniel Severo, Itai Gat, Marton Havasi, Neta Shaul, Peter Holderrieth, Ricky T. Q. Chen, Yaron Lipman","submitted_at":"2024-12-04T17:24:35Z","abstract_excerpt":"The design space of discrete-space diffusion or flow generative models are significantly less well-understood than their continuous-space counterparts, with many works focusing only on a simple masked construction. In this work, we aim to take a holistic approach to the construction of discrete generative models based on continuous-time Markov chains, and for the first time, allow the use of arbitrary discrete probability paths, or colloquially, corruption processes. Through the lens of optimizing the symmetric kinetic energy, we propose velocity formulas that can be applied to any given proba"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.03487","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/2412.03487/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":"2412.03487","created_at":"2026-07-05T09:44:30.188709+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.03487v1","created_at":"2026-07-05T09:44:30.188709+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.03487","created_at":"2026-07-05T09:44:30.188709+00:00"},{"alias_kind":"pith_short_12","alias_value":"DVU6JAHPHJQF","created_at":"2026-07-05T09:44:30.188709+00:00"},{"alias_kind":"pith_short_16","alias_value":"DVU6JAHPHJQFLYK6","created_at":"2026-07-05T09:44:30.188709+00:00"},{"alias_kind":"pith_short_8","alias_value":"DVU6JAHP","created_at":"2026-07-05T09:44:30.188709+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.18518","citing_title":"UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2606.10543","citing_title":"Flexible Flows for Biological Sequence Design","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00773","citing_title":"Accelerating Discrete Diffusion Models with Parallel-In-Time Sampling","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2510.03206","citing_title":"Coevolutionary Continuous Discrete Diffusion: Make Your Diffusion Language Model a Latent Reasoner","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26320","citing_title":"DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matching","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2412.06264","citing_title":"Flow Matching Guide and Code","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08882","citing_title":"Discrete Flow Matching: Convergence Guarantees Under Minimal Assumptions","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09784","citing_title":"Discrete Flow Maps","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06491","citing_title":"Discrete Flow Matching Policy Optimization","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18518","citing_title":"UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH","json":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH.json","graph_json":"https://pith.science/api/pith-number/DVU6JAHPHJQFLYK6T73QSXPYWH/graph.json","events_json":"https://pith.science/api/pith-number/DVU6JAHPHJQFLYK6T73QSXPYWH/events.json","paper":"https://pith.science/paper/DVU6JAHP"},"agent_actions":{"view_html":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH","download_json":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH.json","view_paper":"https://pith.science/paper/DVU6JAHP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.03487&json=true","fetch_graph":"https://pith.science/api/pith-number/DVU6JAHPHJQFLYK6T73QSXPYWH/graph.json","fetch_events":"https://pith.science/api/pith-number/DVU6JAHPHJQFLYK6T73QSXPYWH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH/action/storage_attestation","attest_author":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH/action/author_attestation","sign_citation":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH/action/citation_signature","submit_replication":"https://pith.science/pith/DVU6JAHPHJQFLYK6T73QSXPYWH/action/replication_record"}},"created_at":"2026-07-05T09:44:30.188709+00:00","updated_at":"2026-07-05T09:44:30.188709+00:00"}