{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6QHD4HLDNCFQS5HGBMTGXQI3OZ","short_pith_number":"pith:6QHD4HLD","schema_version":"1.0","canonical_sha256":"f40e3e1d63688b0974e60b266bc11b767aa7a32dd13a6fe5e599d4a3642d3d8c","source":{"kind":"arxiv","id":"2606.30414","version":1},"attestation_state":"computed","paper":{"title":"Diffusion Fine-tuning with Rewarded Moment Matching Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexis Jacq, Arnaud Doucet, Guillaume Couairon, Quentin Berthet, Romuald Elie, Valentin De Bortoli","submitted_at":"2026-06-29T15:00:56Z","abstract_excerpt":"Distillation and Reinforcement Learning (RL) fine-tuning are the primary pillars of diffusion post-training. While traditionally studied in isolation, the interaction between these phases remains poorly understood, and in particular how fine-tuning impacts the generative quality of distilled models. We introduce Rewarded Moment Matching Distillation (RMMD), a novel framework that simultaneously distills diffusion models and maximizes a reward function. RMMD preserves the high-fidelity ``naturalness'' characteristic of advanced distillation (such as 8-step Moment Matching) by adapting the sampl"},"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":"2606.30414","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-29T15:00:56Z","cross_cats_sorted":[],"title_canon_sha256":"5f40772ab54a4737d79c6a6f616eaf0d901cbd53cd8ea679f981df24b093b9cd","abstract_canon_sha256":"f47b53e5b1b30a473a358cbd4b24dd76802d7c160e3707b2aebe04e35fbab84a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:18:14.405882Z","signature_b64":"Jf8ecxOXMUARJI1Rsh8C+cO0T0V69AOOSrjzkGd3MnzGcqhVaJaL8dytKZeZP0Ll/sRhctg7OoOIQn1Xp2f8DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f40e3e1d63688b0974e60b266bc11b767aa7a32dd13a6fe5e599d4a3642d3d8c","last_reissued_at":"2026-06-30T02:18:14.405446Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:18:14.405446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diffusion Fine-tuning with Rewarded Moment Matching Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Alexis Jacq, Arnaud Doucet, Guillaume Couairon, Quentin Berthet, Romuald Elie, Valentin De Bortoli","submitted_at":"2026-06-29T15:00:56Z","abstract_excerpt":"Distillation and Reinforcement Learning (RL) fine-tuning are the primary pillars of diffusion post-training. While traditionally studied in isolation, the interaction between these phases remains poorly understood, and in particular how fine-tuning impacts the generative quality of distilled models. We introduce Rewarded Moment Matching Distillation (RMMD), a novel framework that simultaneously distills diffusion models and maximizes a reward function. RMMD preserves the high-fidelity ``naturalness'' characteristic of advanced distillation (such as 8-step Moment Matching) by adapting the sampl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30414","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/2606.30414/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":"2606.30414","created_at":"2026-06-30T02:18:14.405504+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.30414v1","created_at":"2026-06-30T02:18:14.405504+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30414","created_at":"2026-06-30T02:18:14.405504+00:00"},{"alias_kind":"pith_short_12","alias_value":"6QHD4HLDNCFQ","created_at":"2026-06-30T02:18:14.405504+00:00"},{"alias_kind":"pith_short_16","alias_value":"6QHD4HLDNCFQS5HG","created_at":"2026-06-30T02:18:14.405504+00:00"},{"alias_kind":"pith_short_8","alias_value":"6QHD4HLD","created_at":"2026-06-30T02:18:14.405504+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ","json":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ.json","graph_json":"https://pith.science/api/pith-number/6QHD4HLDNCFQS5HGBMTGXQI3OZ/graph.json","events_json":"https://pith.science/api/pith-number/6QHD4HLDNCFQS5HGBMTGXQI3OZ/events.json","paper":"https://pith.science/paper/6QHD4HLD"},"agent_actions":{"view_html":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ","download_json":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ.json","view_paper":"https://pith.science/paper/6QHD4HLD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.30414&json=true","fetch_graph":"https://pith.science/api/pith-number/6QHD4HLDNCFQS5HGBMTGXQI3OZ/graph.json","fetch_events":"https://pith.science/api/pith-number/6QHD4HLDNCFQS5HGBMTGXQI3OZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ/action/storage_attestation","attest_author":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ/action/author_attestation","sign_citation":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ/action/citation_signature","submit_replication":"https://pith.science/pith/6QHD4HLDNCFQS5HGBMTGXQI3OZ/action/replication_record"}},"created_at":"2026-06-30T02:18:14.405504+00:00","updated_at":"2026-06-30T02:18:14.405504+00:00"}