{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HZQ3N7F7FARI2ANGLVSSDMECNL","short_pith_number":"pith:HZQ3N7F7","schema_version":"1.0","canonical_sha256":"3e61b6fcbf28228d01a65d6521b0826af567196ab435fdb46cd748d7e7515d23","source":{"kind":"arxiv","id":"2607.09753","version":1},"attestation_state":"computed","paper":{"title":"Unified Backbone Refinement for Diffusion Models via Internal-Latent Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Haksoo Lim, Jaesik Choi, Myeongjin Lee, Wonjoon Chang","submitted_at":"2026-07-04T13:24:14Z","abstract_excerpt":"Diffusion models have achieved remarkable success across diverse domains, with performance closely related to the denoising backbones that parameterize the score function. In this paper, we present a systematic, phase-aware analysis of diffusion components and show that abrupt, early-stage fluctuations in deep latents are strongly associated with artifacts. Guided by these findings, we introduce DUNE (Diffusion Unified Network refiNEr), a training-free refinement framework that detects abrupt deviations in deep low-noise internal latents using a shared EMA-based criterion, and applies backbone"},"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":"2607.09753","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-04T13:24:14Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"e2218bffbbeaca4a7be3c93dd050532f94249ce52513febe9a9cd12ca45ea009","abstract_canon_sha256":"f62692b8509d7d19ee96ba04a169f5863f1a19cf092e87ab78231a6302b3ef86"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T00:18:33.907643Z","signature_b64":"yb3OzGwna6kbIRISVTfpJP4P7HCSZFDj2L+GhY8b9RTP9NGq0r9fNyY4TXQMd1ofu5IdLEhUj4v2WtV/Fn3jDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e61b6fcbf28228d01a65d6521b0826af567196ab435fdb46cd748d7e7515d23","last_reissued_at":"2026-07-14T00:18:33.906803Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T00:18:33.906803Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unified Backbone Refinement for Diffusion Models via Internal-Latent Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Haksoo Lim, Jaesik Choi, Myeongjin Lee, Wonjoon Chang","submitted_at":"2026-07-04T13:24:14Z","abstract_excerpt":"Diffusion models have achieved remarkable success across diverse domains, with performance closely related to the denoising backbones that parameterize the score function. In this paper, we present a systematic, phase-aware analysis of diffusion components and show that abrupt, early-stage fluctuations in deep latents are strongly associated with artifacts. Guided by these findings, we introduce DUNE (Diffusion Unified Network refiNEr), a training-free refinement framework that detects abrupt deviations in deep low-noise internal latents using a shared EMA-based criterion, and applies backbone"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.09753","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/2607.09753/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":"2607.09753","created_at":"2026-07-14T00:18:33.907219+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.09753v1","created_at":"2026-07-14T00:18:33.907219+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.09753","created_at":"2026-07-14T00:18:33.907219+00:00"},{"alias_kind":"pith_short_12","alias_value":"HZQ3N7F7FARI","created_at":"2026-07-14T00:18:33.907219+00:00"},{"alias_kind":"pith_short_16","alias_value":"HZQ3N7F7FARI2ANG","created_at":"2026-07-14T00:18:33.907219+00:00"},{"alias_kind":"pith_short_8","alias_value":"HZQ3N7F7","created_at":"2026-07-14T00:18:33.907219+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/HZQ3N7F7FARI2ANGLVSSDMECNL","json":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL.json","graph_json":"https://pith.science/api/pith-number/HZQ3N7F7FARI2ANGLVSSDMECNL/graph.json","events_json":"https://pith.science/api/pith-number/HZQ3N7F7FARI2ANGLVSSDMECNL/events.json","paper":"https://pith.science/paper/HZQ3N7F7"},"agent_actions":{"view_html":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL","download_json":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL.json","view_paper":"https://pith.science/paper/HZQ3N7F7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.09753&json=true","fetch_graph":"https://pith.science/api/pith-number/HZQ3N7F7FARI2ANGLVSSDMECNL/graph.json","fetch_events":"https://pith.science/api/pith-number/HZQ3N7F7FARI2ANGLVSSDMECNL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL/action/storage_attestation","attest_author":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL/action/author_attestation","sign_citation":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL/action/citation_signature","submit_replication":"https://pith.science/pith/HZQ3N7F7FARI2ANGLVSSDMECNL/action/replication_record"}},"created_at":"2026-07-14T00:18:33.907219+00:00","updated_at":"2026-07-14T00:18:33.907219+00:00"}