{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:F2KXQ4WIBH5BZEUORDQAIGMAMF","short_pith_number":"pith:F2KXQ4WI","schema_version":"1.0","canonical_sha256":"2e957872c809fa1c928e88e00419806175f7750f57aee70cd2409eac583268a7","source":{"kind":"arxiv","id":"2505.18445","version":1},"attestation_state":"computed","paper":{"title":"OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cheng Liu, Mike Zheng Shou, Yiren Song","submitted_at":"2025-05-24T01:00:20Z","abstract_excerpt":"Diffusion models have advanced image stylization significantly, yet two core challenges persist: (1) maintaining consistent stylization in complex scenes, particularly identity, composition, and fine details, and (2) preventing style degradation in image-to-image pipelines with style LoRAs. GPT-4o's exceptional stylization consistency highlights the performance gap between open-source methods and proprietary models. To bridge this gap, we propose \\textbf{OmniConsistency}, a universal consistency plugin leveraging large-scale Diffusion Transformers (DiTs). OmniConsistency contributes: (1) an in"},"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":"2505.18445","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-24T01:00:20Z","cross_cats_sorted":[],"title_canon_sha256":"9a1d08d462fa377511e58fe15e1f20b8b70744349d6c31be4c11e11c8a75af72","abstract_canon_sha256":"2c8f15cd07c5b3100f484289961d5d110eec3a2d3299481a353061af8d49200f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:46.584148Z","signature_b64":"T/irpzxwwTAPA5G3Pg6BBnRETUeS83ixCQk/bP+3AKSQDwPiAjKb3nXNJUdbWrMV+qAxBX0r3f5X9Aw8vhcNDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e957872c809fa1c928e88e00419806175f7750f57aee70cd2409eac583268a7","last_reissued_at":"2026-07-05T11:08:46.583718Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:46.583718Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"OmniConsistency: Learning Style-Agnostic Consistency from Paired Stylization Data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cheng Liu, Mike Zheng Shou, Yiren Song","submitted_at":"2025-05-24T01:00:20Z","abstract_excerpt":"Diffusion models have advanced image stylization significantly, yet two core challenges persist: (1) maintaining consistent stylization in complex scenes, particularly identity, composition, and fine details, and (2) preventing style degradation in image-to-image pipelines with style LoRAs. GPT-4o's exceptional stylization consistency highlights the performance gap between open-source methods and proprietary models. To bridge this gap, we propose \\textbf{OmniConsistency}, a universal consistency plugin leveraging large-scale Diffusion Transformers (DiTs). OmniConsistency contributes: (1) an in"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18445","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/2505.18445/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":"2505.18445","created_at":"2026-07-05T11:08:46.583780+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18445v1","created_at":"2026-07-05T11:08:46.583780+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18445","created_at":"2026-07-05T11:08:46.583780+00:00"},{"alias_kind":"pith_short_12","alias_value":"F2KXQ4WIBH5B","created_at":"2026-07-05T11:08:46.583780+00:00"},{"alias_kind":"pith_short_16","alias_value":"F2KXQ4WIBH5BZEUO","created_at":"2026-07-05T11:08:46.583780+00:00"},{"alias_kind":"pith_short_8","alias_value":"F2KXQ4WI","created_at":"2026-07-05T11:08:46.583780+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.14746","citing_title":"Style-CCL: Content-Preserving Style Transfer via Curriculum Continual Learning","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20709","citing_title":"TeleStyle V2: Beyond Content-Preserving Style Transfer with Self-Distillation and Distribution-Matching-Distillation","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01399","citing_title":"PAI-Studio: Cinematic Video Background Replacement with Camera-Aware Motion","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25378","citing_title":"CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17312","citing_title":"VISTA: Triplet-Supervised Video Style Transfer with Diffusion Transformers","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19319","citing_title":"SWEET: Sparse World Modeling with Image Editing for Embodied Task Execution","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17019","citing_title":"StreamingEffect: Real-Time Human-Centric Video Effect Generation","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2602.08392","citing_title":"ST-BiBench: Benchmarking Multi-Stream Multimodal Coordination in Bimanual Embodied Tasks for MLLMs","ref_index":105,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12038","citing_title":"OmniHumanoid: Streaming Cross-Embodiment Video Generation with Paired-Free Adaptation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18993","citing_title":"AutoAWG: Adverse Weather Generation with Adaptive Multi-Controls for Automotive Videos","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF","json":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF.json","graph_json":"https://pith.science/api/pith-number/F2KXQ4WIBH5BZEUORDQAIGMAMF/graph.json","events_json":"https://pith.science/api/pith-number/F2KXQ4WIBH5BZEUORDQAIGMAMF/events.json","paper":"https://pith.science/paper/F2KXQ4WI"},"agent_actions":{"view_html":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF","download_json":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF.json","view_paper":"https://pith.science/paper/F2KXQ4WI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18445&json=true","fetch_graph":"https://pith.science/api/pith-number/F2KXQ4WIBH5BZEUORDQAIGMAMF/graph.json","fetch_events":"https://pith.science/api/pith-number/F2KXQ4WIBH5BZEUORDQAIGMAMF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF/action/storage_attestation","attest_author":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF/action/author_attestation","sign_citation":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF/action/citation_signature","submit_replication":"https://pith.science/pith/F2KXQ4WIBH5BZEUORDQAIGMAMF/action/replication_record"}},"created_at":"2026-07-05T11:08:46.583780+00:00","updated_at":"2026-07-05T11:08:46.583780+00:00"}