{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:27DSFCOMO64KBXA4O22QWB2RCH","short_pith_number":"pith:27DSFCOM","schema_version":"1.0","canonical_sha256":"d7c72289cc77b8a0dc1c76b50b075111fda097ef7df71b46a6ca49076f6500e9","source":{"kind":"arxiv","id":"2407.01606","version":1},"attestation_state":"computed","paper":{"title":"On Discrete Prompt Optimization for Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Boqing Gong, Cho-Jui Hsieh, Ruochen Wang, Ting Liu","submitted_at":"2024-06-27T02:53:01Z","abstract_excerpt":"This paper introduces the first gradient-based framework for prompt optimization in text-to-image diffusion models. We formulate prompt engineering as a discrete optimization problem over the language space. Two major challenges arise in efficiently finding a solution to this problem: (1) Enormous Domain Space: Setting the domain to the entire language space poses significant difficulty to the optimization process. (2) Text Gradient: Efficiently computing the text gradient is challenging, as it requires backpropagating through the inference steps of the diffusion model and a non-differentiable"},"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":"2407.01606","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-27T02:53:01Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV","stat.ML"],"title_canon_sha256":"13f5bb28883d51d4e70f5da17effda568cc8916d6c9a5ae80f5dd132121dd3cf","abstract_canon_sha256":"9540d50119ab765c0f8dc556926020cb11898fa88110a50564eceb1da2c5b9b3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:39:02.686934Z","signature_b64":"5U6LvH1/sDlOLS8ku/rrky/1Z9tIbSsq5Bej61d8D57TcoXNghiyzXGLMRm6T3YfKyY361ijhvy0YIybp0+NDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7c72289cc77b8a0dc1c76b50b075111fda097ef7df71b46a6ca49076f6500e9","last_reissued_at":"2026-07-05T08:39:02.686455Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:39:02.686455Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Discrete Prompt Optimization for Diffusion Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Boqing Gong, Cho-Jui Hsieh, Ruochen Wang, Ting Liu","submitted_at":"2024-06-27T02:53:01Z","abstract_excerpt":"This paper introduces the first gradient-based framework for prompt optimization in text-to-image diffusion models. We formulate prompt engineering as a discrete optimization problem over the language space. Two major challenges arise in efficiently finding a solution to this problem: (1) Enormous Domain Space: Setting the domain to the entire language space poses significant difficulty to the optimization process. (2) Text Gradient: Efficiently computing the text gradient is challenging, as it requires backpropagating through the inference steps of the diffusion model and a non-differentiable"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.01606","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/2407.01606/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":"2407.01606","created_at":"2026-07-05T08:39:02.686515+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.01606v1","created_at":"2026-07-05T08:39:02.686515+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.01606","created_at":"2026-07-05T08:39:02.686515+00:00"},{"alias_kind":"pith_short_12","alias_value":"27DSFCOMO64K","created_at":"2026-07-05T08:39:02.686515+00:00"},{"alias_kind":"pith_short_16","alias_value":"27DSFCOMO64KBXA4","created_at":"2026-07-05T08:39:02.686515+00:00"},{"alias_kind":"pith_short_8","alias_value":"27DSFCOM","created_at":"2026-07-05T08:39:02.686515+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.15803","citing_title":"Embedding-perturbed Exploration Preference Optimization for Flow Models","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07907","citing_title":"Consistency Regularised Gradient Flows for Inverse Problems","ref_index":240,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH","json":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH.json","graph_json":"https://pith.science/api/pith-number/27DSFCOMO64KBXA4O22QWB2RCH/graph.json","events_json":"https://pith.science/api/pith-number/27DSFCOMO64KBXA4O22QWB2RCH/events.json","paper":"https://pith.science/paper/27DSFCOM"},"agent_actions":{"view_html":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH","download_json":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH.json","view_paper":"https://pith.science/paper/27DSFCOM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.01606&json=true","fetch_graph":"https://pith.science/api/pith-number/27DSFCOMO64KBXA4O22QWB2RCH/graph.json","fetch_events":"https://pith.science/api/pith-number/27DSFCOMO64KBXA4O22QWB2RCH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH/action/storage_attestation","attest_author":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH/action/author_attestation","sign_citation":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH/action/citation_signature","submit_replication":"https://pith.science/pith/27DSFCOMO64KBXA4O22QWB2RCH/action/replication_record"}},"created_at":"2026-07-05T08:39:02.686515+00:00","updated_at":"2026-07-05T08:39:02.686515+00:00"}