{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HXK2XIZCBXQFZAADARAHVZDVCQ","short_pith_number":"pith:HXK2XIZC","schema_version":"1.0","canonical_sha256":"3dd5aba3220de05c800304407ae475142df11a9a89f13ff6aa8340ac264d6bfd","source":{"kind":"arxiv","id":"2507.11554","version":4},"attestation_state":"computed","paper":{"title":"Inversion-DPO: Precise and Efficient Post-Training for Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Changyuan Yang, Chenye Meng, Guang Yang, Lingyun Sun, Shengyuan Zhang, Yang Ling, Yize Li, Zejian Li, Zhiyuan Yang, Zhongni Liu","submitted_at":"2025-07-14T02:59:28Z","abstract_excerpt":"Recent advancements in diffusion models (DMs) have been propelled by alignment methods that post-train models to better conform to human preferences. However, these approaches typically require computation-intensive training of a base model and a reward model, which not only incurs substantial computational overhead but may also compromise model accuracy and training efficiency. To address these limitations, we propose Inversion-DPO, a novel alignment framework that circumvents reward modeling by reformulating Direct Preference Optimization (DPO) with DDIM inversion for DMs. Our method conduct"},"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":"2507.11554","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-14T02:59:28Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5bdf647b2663ea69aff616a06ff5efc670eda2073dc1ac10b8db31417ec46ece","abstract_canon_sha256":"8b474ef4549586e28eac4b75b681dc40fd13ed5b4f0844398ef02a640963e295"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:47:26.144394Z","signature_b64":"zNAhR38Mz3weZkeT6NZtQPunrCMntZ6Yp9v+kFrjtGbD2SUyQ6xhWL0KjaqJZHmJ4mCz6lJ7MUg2w9nia3CxAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3dd5aba3220de05c800304407ae475142df11a9a89f13ff6aa8340ac264d6bfd","last_reissued_at":"2026-07-05T11:47:26.143835Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:47:26.143835Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Inversion-DPO: Precise and Efficient Post-Training for Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Changyuan Yang, Chenye Meng, Guang Yang, Lingyun Sun, Shengyuan Zhang, Yang Ling, Yize Li, Zejian Li, Zhiyuan Yang, Zhongni Liu","submitted_at":"2025-07-14T02:59:28Z","abstract_excerpt":"Recent advancements in diffusion models (DMs) have been propelled by alignment methods that post-train models to better conform to human preferences. However, these approaches typically require computation-intensive training of a base model and a reward model, which not only incurs substantial computational overhead but may also compromise model accuracy and training efficiency. To address these limitations, we propose Inversion-DPO, a novel alignment framework that circumvents reward modeling by reformulating Direct Preference Optimization (DPO) with DDIM inversion for DMs. Our method conduct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.11554","kind":"arxiv","version":4},"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/2507.11554/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":"2507.11554","created_at":"2026-07-05T11:47:26.143893+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.11554v4","created_at":"2026-07-05T11:47:26.143893+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.11554","created_at":"2026-07-05T11:47:26.143893+00:00"},{"alias_kind":"pith_short_12","alias_value":"HXK2XIZCBXQF","created_at":"2026-07-05T11:47:26.143893+00:00"},{"alias_kind":"pith_short_16","alias_value":"HXK2XIZCBXQFZAAD","created_at":"2026-07-05T11:47:26.143893+00:00"},{"alias_kind":"pith_short_8","alias_value":"HXK2XIZC","created_at":"2026-07-05T11:47:26.143893+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/HXK2XIZCBXQFZAADARAHVZDVCQ","json":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ.json","graph_json":"https://pith.science/api/pith-number/HXK2XIZCBXQFZAADARAHVZDVCQ/graph.json","events_json":"https://pith.science/api/pith-number/HXK2XIZCBXQFZAADARAHVZDVCQ/events.json","paper":"https://pith.science/paper/HXK2XIZC"},"agent_actions":{"view_html":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ","download_json":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ.json","view_paper":"https://pith.science/paper/HXK2XIZC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.11554&json=true","fetch_graph":"https://pith.science/api/pith-number/HXK2XIZCBXQFZAADARAHVZDVCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/HXK2XIZCBXQFZAADARAHVZDVCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ/action/storage_attestation","attest_author":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ/action/author_attestation","sign_citation":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ/action/citation_signature","submit_replication":"https://pith.science/pith/HXK2XIZCBXQFZAADARAHVZDVCQ/action/replication_record"}},"created_at":"2026-07-05T11:47:26.143893+00:00","updated_at":"2026-07-05T11:47:26.143893+00:00"}