{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TGAVJMWCNBNOKWI5E5PWGXQMSY","short_pith_number":"pith:TGAVJMWC","schema_version":"1.0","canonical_sha256":"998154b2c2685ae5591d275f635e0c962ff5538cbe96d9b30e584a418e3894b4","source":{"kind":"arxiv","id":"2310.10012","version":4},"attestation_state":"computed","paper":{"title":"Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Li, Chia-Mu Yu, Chia-Yi Hsu, Chih-Hsun Lin, Chulin Xie, Chun-Ying Huang, Jia-You Chen, Pin-Yu Chen, Yu-Lin Tsai","submitted_at":"2023-10-16T02:11:20Z","abstract_excerpt":"Diffusion models for text-to-image (T2I) synthesis, such as Stable Diffusion (SD), have recently demonstrated exceptional capabilities for generating high-quality content. However, this progress has raised several concerns of potential misuse, particularly in creating copyrighted, prohibited, and restricted content, or NSFW (not safe for work) images. While efforts have been made to mitigate such problems, either by implementing a safety filter at the evaluation stage or by fine-tuning models to eliminate undesirable concepts or styles, the effectiveness of these safety measures in dealing wit"},"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":"2310.10012","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-10-16T02:11:20Z","cross_cats_sorted":[],"title_canon_sha256":"4ef047bdf321f2c01d22d9c79235fa8e9b385b7b88023586e3ae6bd986b48753","abstract_canon_sha256":"963fc8b8cbe6a66eb937ec92b1f22328c3678809c92b66e590882cce27ebc138"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:34.323749Z","signature_b64":"rjDO1nweOYlTreeaAJ0U5NNmOgwC995orulknvYq+nm0aJyAVrezMSMRkBoophNAHIDohJCMEiNET0YKbYegAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"998154b2c2685ae5591d275f635e0c962ff5538cbe96d9b30e584a418e3894b4","last_reissued_at":"2026-07-05T08:28:34.323300Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:34.323300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bo Li, Chia-Mu Yu, Chia-Yi Hsu, Chih-Hsun Lin, Chulin Xie, Chun-Ying Huang, Jia-You Chen, Pin-Yu Chen, Yu-Lin Tsai","submitted_at":"2023-10-16T02:11:20Z","abstract_excerpt":"Diffusion models for text-to-image (T2I) synthesis, such as Stable Diffusion (SD), have recently demonstrated exceptional capabilities for generating high-quality content. However, this progress has raised several concerns of potential misuse, particularly in creating copyrighted, prohibited, and restricted content, or NSFW (not safe for work) images. While efforts have been made to mitigate such problems, either by implementing a safety filter at the evaluation stage or by fine-tuning models to eliminate undesirable concepts or styles, the effectiveness of these safety measures in dealing wit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.10012","kind":"arxiv","version":4},"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/2310.10012/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":"2310.10012","created_at":"2026-07-05T08:28:34.323363+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.10012v4","created_at":"2026-07-05T08:28:34.323363+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.10012","created_at":"2026-07-05T08:28:34.323363+00:00"},{"alias_kind":"pith_short_12","alias_value":"TGAVJMWCNBNO","created_at":"2026-07-05T08:28:34.323363+00:00"},{"alias_kind":"pith_short_16","alias_value":"TGAVJMWCNBNOKWI5","created_at":"2026-07-05T08:28:34.323363+00:00"},{"alias_kind":"pith_short_8","alias_value":"TGAVJMWC","created_at":"2026-07-05T08:28:34.323363+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24192","citing_title":"Co-occurring associated retained concepts in Diffusion Unlearning","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23267","citing_title":"Safe Few-Step Generation via Velocity Editing","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06875","citing_title":"Unified Safe In-context Image Generation in Multimodal Diffusion Transformers via Restricting Unsafe Information Flows","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00402","citing_title":"The Illusion of High Utility in Safety Alignment of Text-to-Image Diffusion Models","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01658","citing_title":"CoreUnlearn: Rethinking Concept Unlearning through Disentangled Component-Level Erasure in Text-guided Diffusion Models","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19739","citing_title":"FlowErase-RL: Rethinking Concept Erasure as Reward Optimization in Flow Matching Models","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26332","citing_title":"Erased but Exploitable: Black-box Embedding-Aware Prompting Against Unlearned Text-to-Image Diffusion Models","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01481","citing_title":"SafeGen-Bench: Benchmarking Safety in Image-Conditioned Text-to-Video Generation","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19739","citing_title":"FlowErase-RL: Rethinking Concept Erasure as Reward Optimization in Flow Matching Models","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2602.00616","citing_title":"SPOT: Selective Prompt Projection via Total Variation for Inference-Only Safe Text-to-Image Generation","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10180","citing_title":"What Concepts Lie Within? Detecting and Suppressing Risky Content in Diffusion Transformers","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10198","citing_title":"Empty SPACE: Cross-Attention Sparsity for Concept Erasure in Diffusion Models","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09405","citing_title":"EGLOCE: Training-Free Energy-Guided Latent Optimization for Concept Erasure","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15829","citing_title":"Beyond Text Prompts: Precise Concept Erasure through Text-Image Collaboration","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01761","citing_title":"TrajShield: Trajectory-Level Safety Mediation for Defending Text-to-Video Models Against Jailbreak Attacks","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY","json":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY.json","graph_json":"https://pith.science/api/pith-number/TGAVJMWCNBNOKWI5E5PWGXQMSY/graph.json","events_json":"https://pith.science/api/pith-number/TGAVJMWCNBNOKWI5E5PWGXQMSY/events.json","paper":"https://pith.science/paper/TGAVJMWC"},"agent_actions":{"view_html":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY","download_json":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY.json","view_paper":"https://pith.science/paper/TGAVJMWC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.10012&json=true","fetch_graph":"https://pith.science/api/pith-number/TGAVJMWCNBNOKWI5E5PWGXQMSY/graph.json","fetch_events":"https://pith.science/api/pith-number/TGAVJMWCNBNOKWI5E5PWGXQMSY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY/action/storage_attestation","attest_author":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY/action/author_attestation","sign_citation":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY/action/citation_signature","submit_replication":"https://pith.science/pith/TGAVJMWCNBNOKWI5E5PWGXQMSY/action/replication_record"}},"created_at":"2026-07-05T08:28:34.323363+00:00","updated_at":"2026-07-05T08:28:34.323363+00:00"}