{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UCUNOSANJ6PSYLKTACP2N33E5M","short_pith_number":"pith:UCUNOSAN","schema_version":"1.0","canonical_sha256":"a0a8d7480d4f9f2c2d53009fa6ef64eb06d8f81be03ff05eb7aa47f25e30773d","source":{"kind":"arxiv","id":"2504.00457","version":3},"attestation_state":"computed","paper":{"title":"Distilling Multi-view Diffusion Models into 3D Generators","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hao Qin, Luyuan Chen, Mengxu Lu, Ming Kong, Qiang Zhu","submitted_at":"2025-04-01T06:32:48Z","abstract_excerpt":"We introduce DD3G, a formulation that Distills a multi-view Diffusion model (MV-DM) into a 3D Generator using gaussian splatting. DD3G compresses and integrates extensive visual and spatial geometric knowledge from the MV-DM by simulating its ordinary differential equation (ODE) trajectory, ensuring the distilled generator generalizes better than those trained solely on 3D data. Unlike previous amortized optimization approaches, we align the MV-DM and 3D generator representation spaces to transfer the teacher's probabilistic flow to the student, thus avoiding inconsistencies in optimization ob"},"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":"2504.00457","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-04-01T06:32:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"dd43216c349d1edd088ab4378c414cdba6b7e1db7e02d2eb0142899c3d21d39c","abstract_canon_sha256":"883a5a6502bd3e945b219aed00be7035fd9832e284dcd1666227b7111e46c69d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:43:47.233662Z","signature_b64":"z2lecNwtDWyw522jJax6a/cdBdx1yjjTfqtHyBFBA8smWrzWz8a4JmjFnFv/6MG50XYEbqZb7CCs/WdKZPN1Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a0a8d7480d4f9f2c2d53009fa6ef64eb06d8f81be03ff05eb7aa47f25e30773d","last_reissued_at":"2026-07-05T10:43:47.233177Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:43:47.233177Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distilling Multi-view Diffusion Models into 3D Generators","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Hao Qin, Luyuan Chen, Mengxu Lu, Ming Kong, Qiang Zhu","submitted_at":"2025-04-01T06:32:48Z","abstract_excerpt":"We introduce DD3G, a formulation that Distills a multi-view Diffusion model (MV-DM) into a 3D Generator using gaussian splatting. DD3G compresses and integrates extensive visual and spatial geometric knowledge from the MV-DM by simulating its ordinary differential equation (ODE) trajectory, ensuring the distilled generator generalizes better than those trained solely on 3D data. Unlike previous amortized optimization approaches, we align the MV-DM and 3D generator representation spaces to transfer the teacher's probabilistic flow to the student, thus avoiding inconsistencies in optimization ob"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.00457","kind":"arxiv","version":3},"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/2504.00457/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":"2504.00457","created_at":"2026-07-05T10:43:47.233233+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.00457v3","created_at":"2026-07-05T10:43:47.233233+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.00457","created_at":"2026-07-05T10:43:47.233233+00:00"},{"alias_kind":"pith_short_12","alias_value":"UCUNOSANJ6PS","created_at":"2026-07-05T10:43:47.233233+00:00"},{"alias_kind":"pith_short_16","alias_value":"UCUNOSANJ6PSYLKT","created_at":"2026-07-05T10:43:47.233233+00:00"},{"alias_kind":"pith_short_8","alias_value":"UCUNOSAN","created_at":"2026-07-05T10:43:47.233233+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.05162","citing_title":"Controllable Dynamic 3D Shape Generation via 3D Trajectories and Text","ref_index":34,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M","json":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M.json","graph_json":"https://pith.science/api/pith-number/UCUNOSANJ6PSYLKTACP2N33E5M/graph.json","events_json":"https://pith.science/api/pith-number/UCUNOSANJ6PSYLKTACP2N33E5M/events.json","paper":"https://pith.science/paper/UCUNOSAN"},"agent_actions":{"view_html":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M","download_json":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M.json","view_paper":"https://pith.science/paper/UCUNOSAN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.00457&json=true","fetch_graph":"https://pith.science/api/pith-number/UCUNOSANJ6PSYLKTACP2N33E5M/graph.json","fetch_events":"https://pith.science/api/pith-number/UCUNOSANJ6PSYLKTACP2N33E5M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M/action/storage_attestation","attest_author":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M/action/author_attestation","sign_citation":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M/action/citation_signature","submit_replication":"https://pith.science/pith/UCUNOSANJ6PSYLKTACP2N33E5M/action/replication_record"}},"created_at":"2026-07-05T10:43:47.233233+00:00","updated_at":"2026-07-05T10:43:47.233233+00:00"}