{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3RLUUPYLVUX53C6MMSYPWIKA5E","short_pith_number":"pith:3RLUUPYL","schema_version":"1.0","canonical_sha256":"dc574a3f0bad2fdd8bcc64b0fb2140e91a4b222e04f64c661bdfb2485b303ed9","source":{"kind":"arxiv","id":"2402.05841","version":2},"attestation_state":"computed","paper":{"title":"Dirichlet Flow Matching with Applications to DNA Sequence Design","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.BM","authors_text":"Bonnie Berger, Bowen Jing, Chenyu Wang, Gabriele Corso, Hannes Stark, Regina Barzilay, Tommi Jaakkola","submitted_at":"2024-02-08T17:18:01Z","abstract_excerpt":"Discrete diffusion or flow models could enable faster and more controllable sequence generation than autoregressive models. We show that na\\\"ive linear flow matching on the simplex is insufficient toward this goal since it suffers from discontinuities in the training target and further pathologies. To overcome this, we develop Dirichlet flow matching on the simplex based on mixtures of Dirichlet distributions as probability paths. In this framework, we derive a connection between the mixtures' scores and the flow's vector field that allows for classifier and classifier-free guidance. 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":"2402.05841","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-bio.BM","submitted_at":"2024-02-08T17:18:01Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"06a815fb956b217281ad21305281cb5659f1f882b4c170e13f64408656ff9bdd","abstract_canon_sha256":"be9f5e1e57dad902cde3bdfeee6f1b82c494be2f2d321a365acb89d3b93993fd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:25:31.562680Z","signature_b64":"TEg80uY9qvDEfBmnBMlu9+Vj/oKRA4YyeKYw/6LCFlLz6Apkv/0UHQNfomIT0E5ft5nN/2ucz2T/4+Oujbs9Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dc574a3f0bad2fdd8bcc64b0fb2140e91a4b222e04f64c661bdfb2485b303ed9","last_reissued_at":"2026-07-05T08:25:31.562184Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:25:31.562184Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dirichlet Flow Matching with Applications to DNA Sequence Design","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.BM","authors_text":"Bonnie Berger, Bowen Jing, Chenyu Wang, Gabriele Corso, Hannes Stark, Regina Barzilay, Tommi Jaakkola","submitted_at":"2024-02-08T17:18:01Z","abstract_excerpt":"Discrete diffusion or flow models could enable faster and more controllable sequence generation than autoregressive models. We show that na\\\"ive linear flow matching on the simplex is insufficient toward this goal since it suffers from discontinuities in the training target and further pathologies. To overcome this, we develop Dirichlet flow matching on the simplex based on mixtures of Dirichlet distributions as probability paths. In this framework, we derive a connection between the mixtures' scores and the flow's vector field that allows for classifier and classifier-free guidance. Further, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.05841","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/2402.05841/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":"2402.05841","created_at":"2026-07-05T08:25:31.562247+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.05841v2","created_at":"2026-07-05T08:25:31.562247+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.05841","created_at":"2026-07-05T08:25:31.562247+00:00"},{"alias_kind":"pith_short_12","alias_value":"3RLUUPYLVUX5","created_at":"2026-07-05T08:25:31.562247+00:00"},{"alias_kind":"pith_short_16","alias_value":"3RLUUPYLVUX53C6M","created_at":"2026-07-05T08:25:31.562247+00:00"},{"alias_kind":"pith_short_8","alias_value":"3RLUUPYL","created_at":"2026-07-05T08:25:31.562247+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.10543","citing_title":"Flexible Flows for Biological Sequence Design","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08501","citing_title":"Back on Track: Aligning Rewards and States for Reasoning in Diffusion Large Language Models","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21831","citing_title":"Generative Modeling of Discrete Data Using Geometric Latent Subspaces","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05629","citing_title":"Spherical Flows for Sampling Categorical Data","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05629","citing_title":"Spherical Flows for Sampling Categorical Data","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07193","citing_title":"Coupling Models for One-Step Discrete Generation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17310","citing_title":"Interpolating Discrete Diffusion Models with Controllable Resampling","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E","json":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E.json","graph_json":"https://pith.science/api/pith-number/3RLUUPYLVUX53C6MMSYPWIKA5E/graph.json","events_json":"https://pith.science/api/pith-number/3RLUUPYLVUX53C6MMSYPWIKA5E/events.json","paper":"https://pith.science/paper/3RLUUPYL"},"agent_actions":{"view_html":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E","download_json":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E.json","view_paper":"https://pith.science/paper/3RLUUPYL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.05841&json=true","fetch_graph":"https://pith.science/api/pith-number/3RLUUPYLVUX53C6MMSYPWIKA5E/graph.json","fetch_events":"https://pith.science/api/pith-number/3RLUUPYLVUX53C6MMSYPWIKA5E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E/action/storage_attestation","attest_author":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E/action/author_attestation","sign_citation":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E/action/citation_signature","submit_replication":"https://pith.science/pith/3RLUUPYLVUX53C6MMSYPWIKA5E/action/replication_record"}},"created_at":"2026-07-05T08:25:31.562247+00:00","updated_at":"2026-07-05T08:25:31.562247+00:00"}