{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5AO7W24Z6SHHTTXONZV2MRRRSE","short_pith_number":"pith:5AO7W24Z","schema_version":"1.0","canonical_sha256":"e81dfb6b99f48e79ceee6e6ba64631913be9bc5d6c9fce531789b9e5123b13ce","source":{"kind":"arxiv","id":"2504.11850","version":1},"attestation_state":"computed","paper":{"title":"ACE: Attentional Concept Erasure in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Finn Carter","submitted_at":"2025-04-16T08:16:28Z","abstract_excerpt":"Large text-to-image diffusion models have demonstrated remarkable image synthesis capabilities, but their indiscriminate training on Internet-scale data has led to learned concepts that enable harmful, copyrighted, or otherwise undesirable content generation. We address the task of concept erasure in diffusion models, i.e., removing a specified concept from a pre-trained model such that prompting the concept (or related synonyms) no longer yields its depiction, while preserving the model's ability to generate other content. We propose a novel method, Attentional Concept Erasure (ACE), that int"},"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":"2504.11850","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-04-16T08:16:28Z","cross_cats_sorted":[],"title_canon_sha256":"556201cfab14972b2bc9939c264802437b381e6dce3fa7855dc34c7b65e96efc","abstract_canon_sha256":"c5be688c82b89cd2b76450a85a6c851e16a08a7b104faf1fef8075776d92d8ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:49:58.700836Z","signature_b64":"rqzPcsBspeWGogIDsDs1zZ0Om61flm77OXQzrYcc+QJQmiQQAsrx+egiizipzJ1bjqWKaadyMqenQ90PhtTyDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e81dfb6b99f48e79ceee6e6ba64631913be9bc5d6c9fce531789b9e5123b13ce","last_reissued_at":"2026-07-05T10:49:58.700357Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:49:58.700357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ACE: Attentional Concept Erasure in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Finn Carter","submitted_at":"2025-04-16T08:16:28Z","abstract_excerpt":"Large text-to-image diffusion models have demonstrated remarkable image synthesis capabilities, but their indiscriminate training on Internet-scale data has led to learned concepts that enable harmful, copyrighted, or otherwise undesirable content generation. We address the task of concept erasure in diffusion models, i.e., removing a specified concept from a pre-trained model such that prompting the concept (or related synonyms) no longer yields its depiction, while preserving the model's ability to generate other content. We propose a novel method, Attentional Concept Erasure (ACE), that int"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.11850","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/2504.11850/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":"2504.11850","created_at":"2026-07-05T10:49:58.700418+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.11850v1","created_at":"2026-07-05T10:49:58.700418+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.11850","created_at":"2026-07-05T10:49:58.700418+00:00"},{"alias_kind":"pith_short_12","alias_value":"5AO7W24Z6SHH","created_at":"2026-07-05T10:49:58.700418+00:00"},{"alias_kind":"pith_short_16","alias_value":"5AO7W24Z6SHHTTXO","created_at":"2026-07-05T10:49:58.700418+00:00"},{"alias_kind":"pith_short_8","alias_value":"5AO7W24Z","created_at":"2026-07-05T10:49:58.700418+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.12968","citing_title":"GrOCE:Graph-Guided Online Concept Erasure for Text-to-Image Diffusion Models","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE","json":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE.json","graph_json":"https://pith.science/api/pith-number/5AO7W24Z6SHHTTXONZV2MRRRSE/graph.json","events_json":"https://pith.science/api/pith-number/5AO7W24Z6SHHTTXONZV2MRRRSE/events.json","paper":"https://pith.science/paper/5AO7W24Z"},"agent_actions":{"view_html":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE","download_json":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE.json","view_paper":"https://pith.science/paper/5AO7W24Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.11850&json=true","fetch_graph":"https://pith.science/api/pith-number/5AO7W24Z6SHHTTXONZV2MRRRSE/graph.json","fetch_events":"https://pith.science/api/pith-number/5AO7W24Z6SHHTTXONZV2MRRRSE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE/action/storage_attestation","attest_author":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE/action/author_attestation","sign_citation":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE/action/citation_signature","submit_replication":"https://pith.science/pith/5AO7W24Z6SHHTTXONZV2MRRRSE/action/replication_record"}},"created_at":"2026-07-05T10:49:58.700418+00:00","updated_at":"2026-07-05T10:49:58.700418+00:00"}