{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:YR6D5VO7XYF5D4PNXX6SST4NUU","short_pith_number":"pith:YR6D5VO7","schema_version":"1.0","canonical_sha256":"c47c3ed5dfbe0bd1f1edbdfd294f8da52a80682286139134911b5ea087628a77","source":{"kind":"arxiv","id":"2412.20413","version":2},"attestation_state":"computed","paper":{"title":"EraseAnything: Enabling Concept Erasure in Rectified Flow Transformers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bang Zhang, Daiheng Gao, Jiaming Chu, Jian Zhao, Jie Zhang, Mengxi Jia, Shaw Walters, Shilin Lu, Weiming Zhang, Wenbo Zhou, Zhaoxin Fan","submitted_at":"2024-12-29T09:42:53Z","abstract_excerpt":"Removing unwanted concepts from large-scale text-to-image (T2I) diffusion models while maintaining their overall generative quality remains an open challenge. This difficulty is especially pronounced in emerging paradigms, such as Stable Diffusion (SD) v3 and Flux, which incorporate flow matching and transformer-based architectures. These advancements limit the transferability of existing concept-erasure techniques that were originally designed for the previous T2I paradigm (e.g., SD v1.4). In this work, we introduce EraseAnything, the first method specifically developed to address concept era"},"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":"2412.20413","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-29T09:42:53Z","cross_cats_sorted":[],"title_canon_sha256":"48795e9c7e33476a927a42eba1bab6524b5af5c7e794af62e29484c9912d7e5f","abstract_canon_sha256":"81209ca7b98b725adab4008ccf8cac87c233e341ab6755209140205d26a37353"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:56:05.002526Z","signature_b64":"O6d4uJjQpFCQKVDTMfeMGBnosnSX78MJsl7wMpkHXcQIxsFkyozDRVLjY0GT9Ee0hhHFpAq9ATuzp/jA/Kf0AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c47c3ed5dfbe0bd1f1edbdfd294f8da52a80682286139134911b5ea087628a77","last_reissued_at":"2026-07-05T09:56:05.002104Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:56:05.002104Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EraseAnything: Enabling Concept Erasure in Rectified Flow Transformers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bang Zhang, Daiheng Gao, Jiaming Chu, Jian Zhao, Jie Zhang, Mengxi Jia, Shaw Walters, Shilin Lu, Weiming Zhang, Wenbo Zhou, Zhaoxin Fan","submitted_at":"2024-12-29T09:42:53Z","abstract_excerpt":"Removing unwanted concepts from large-scale text-to-image (T2I) diffusion models while maintaining their overall generative quality remains an open challenge. This difficulty is especially pronounced in emerging paradigms, such as Stable Diffusion (SD) v3 and Flux, which incorporate flow matching and transformer-based architectures. These advancements limit the transferability of existing concept-erasure techniques that were originally designed for the previous T2I paradigm (e.g., SD v1.4). In this work, we introduce EraseAnything, the first method specifically developed to address concept era"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.20413","kind":"arxiv","version":2},"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/2412.20413/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":"2412.20413","created_at":"2026-07-05T09:56:05.002170+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.20413v2","created_at":"2026-07-05T09:56:05.002170+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.20413","created_at":"2026-07-05T09:56:05.002170+00:00"},{"alias_kind":"pith_short_12","alias_value":"YR6D5VO7XYF5","created_at":"2026-07-05T09:56:05.002170+00:00"},{"alias_kind":"pith_short_16","alias_value":"YR6D5VO7XYF5D4PN","created_at":"2026-07-05T09:56:05.002170+00:00"},{"alias_kind":"pith_short_8","alias_value":"YR6D5VO7","created_at":"2026-07-05T09:56:05.002170+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11689","citing_title":"RankVR: Low-Rank Structure Perception and Value Recalibration for Robust Composed Image Retrieval","ref_index":106,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08144","citing_title":"IMAGINE: Adaptive Schema-Imagery Enhanced Composition for Composed Video Retrieval","ref_index":94,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09901","citing_title":"On the Controllability-Fidelity Frontier in Diffusion Editing","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17898","citing_title":"ReTrack: Evidence-Driven Dual-Stream Directional Anchor Calibration Network for Composed Video Retrieval","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18051","citing_title":"INTENT: Invariance and Discrimination-aware Noise Mitigation for Robust Composed Image Retrieval","ref_index":90,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18047","citing_title":"GS-STVSR: Ultra-Efficient Continuous Spatio-Temporal Video Super-Resolution via 2D Gaussian Splatting","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18037","citing_title":"HABIT: Chrono-Synergia Robust Progressive Learning Framework for Composed Image Retrieval","ref_index":78,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU","json":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU.json","graph_json":"https://pith.science/api/pith-number/YR6D5VO7XYF5D4PNXX6SST4NUU/graph.json","events_json":"https://pith.science/api/pith-number/YR6D5VO7XYF5D4PNXX6SST4NUU/events.json","paper":"https://pith.science/paper/YR6D5VO7"},"agent_actions":{"view_html":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU","download_json":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU.json","view_paper":"https://pith.science/paper/YR6D5VO7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.20413&json=true","fetch_graph":"https://pith.science/api/pith-number/YR6D5VO7XYF5D4PNXX6SST4NUU/graph.json","fetch_events":"https://pith.science/api/pith-number/YR6D5VO7XYF5D4PNXX6SST4NUU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU/action/storage_attestation","attest_author":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU/action/author_attestation","sign_citation":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU/action/citation_signature","submit_replication":"https://pith.science/pith/YR6D5VO7XYF5D4PNXX6SST4NUU/action/replication_record"}},"created_at":"2026-07-05T09:56:05.002170+00:00","updated_at":"2026-07-05T09:56:05.002170+00:00"}