{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EDM3BBXJXBMEW5PVGKIVRFNWMP","short_pith_number":"pith:EDM3BBXJ","schema_version":"1.0","canonical_sha256":"20d9b086e9b8584b75f532915895b663c3ef20fa26728e4cd6dbbb3ec073e83d","source":{"kind":"arxiv","id":"2403.19103","version":4},"attestation_state":"computed","paper":{"title":"Automated Black-box Prompt Engineering for Personalized Text-to-Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander Robey, George J. Pappas, Hamed Hassani, Joshua Nathaniel Williams, J. Zico Kolter, Naoki Murata, Ruslan Salakhutdinov, Yiding Jiang, Yuki Mitsufuji, Yutong He","submitted_at":"2024-03-28T02:35:53Z","abstract_excerpt":"Prompt engineering is an effective but labor-intensive way to control text-to-image (T2I) generative models. Its time-intensive nature and complexity have spurred the development of algorithms for automated prompt generation. However, these methods often struggle with transferability across T2I models, require white-box access to the underlying model, or produce non-intuitive prompts. In this work, we introduce PRISM, an algorithm that automatically produces human-interpretable and transferable prompts that can effectively generate desired concepts given only black-box access to T2I models. 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":"2403.19103","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-03-28T02:35:53Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"4d81b28689dd08d58bb04e4d843ee0375c1c5c6695c888315c70dd2aaaf47e2f","abstract_canon_sha256":"0840ed9292ea80eb3ff3bbc37f86342b9c1a0bea1f56b6340d189313715ba551"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:39.433967Z","signature_b64":"SxSl+e3n8TuyhJfnfMqANBLi5a0gQDysBD/iF2lAWBJjyd5Mh10QsPzEXl6CG5X0ju2r9HeWysY6n6gzYsRfAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20d9b086e9b8584b75f532915895b663c3ef20fa26728e4cd6dbbb3ec073e83d","last_reissued_at":"2026-07-05T11:54:39.433496Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:39.433496Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Automated Black-box Prompt Engineering for Personalized Text-to-Image Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander Robey, George J. Pappas, Hamed Hassani, Joshua Nathaniel Williams, J. Zico Kolter, Naoki Murata, Ruslan Salakhutdinov, Yiding Jiang, Yuki Mitsufuji, Yutong He","submitted_at":"2024-03-28T02:35:53Z","abstract_excerpt":"Prompt engineering is an effective but labor-intensive way to control text-to-image (T2I) generative models. Its time-intensive nature and complexity have spurred the development of algorithms for automated prompt generation. However, these methods often struggle with transferability across T2I models, require white-box access to the underlying model, or produce non-intuitive prompts. In this work, we introduce PRISM, an algorithm that automatically produces human-interpretable and transferable prompts that can effectively generate desired concepts given only black-box access to T2I models. In"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.19103","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/2403.19103/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":"2403.19103","created_at":"2026-07-05T11:54:39.433555+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.19103v4","created_at":"2026-07-05T11:54:39.433555+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.19103","created_at":"2026-07-05T11:54:39.433555+00:00"},{"alias_kind":"pith_short_12","alias_value":"EDM3BBXJXBME","created_at":"2026-07-05T11:54:39.433555+00:00"},{"alias_kind":"pith_short_16","alias_value":"EDM3BBXJXBMEW5PV","created_at":"2026-07-05T11:54:39.433555+00:00"},{"alias_kind":"pith_short_8","alias_value":"EDM3BBXJ","created_at":"2026-07-05T11:54:39.433555+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.06061","citing_title":"PromptEvolver: Prompt Inversion through Evolutionary Optimization in Natural-Language Space","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP","json":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP.json","graph_json":"https://pith.science/api/pith-number/EDM3BBXJXBMEW5PVGKIVRFNWMP/graph.json","events_json":"https://pith.science/api/pith-number/EDM3BBXJXBMEW5PVGKIVRFNWMP/events.json","paper":"https://pith.science/paper/EDM3BBXJ"},"agent_actions":{"view_html":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP","download_json":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP.json","view_paper":"https://pith.science/paper/EDM3BBXJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.19103&json=true","fetch_graph":"https://pith.science/api/pith-number/EDM3BBXJXBMEW5PVGKIVRFNWMP/graph.json","fetch_events":"https://pith.science/api/pith-number/EDM3BBXJXBMEW5PVGKIVRFNWMP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP/action/storage_attestation","attest_author":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP/action/author_attestation","sign_citation":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP/action/citation_signature","submit_replication":"https://pith.science/pith/EDM3BBXJXBMEW5PVGKIVRFNWMP/action/replication_record"}},"created_at":"2026-07-05T11:54:39.433555+00:00","updated_at":"2026-07-05T11:54:39.433555+00:00"}