{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:UDTQA4TGDULJB3DVI5H3MTYQM5","short_pith_number":"pith:UDTQA4TG","schema_version":"1.0","canonical_sha256":"a0e70072661d1690ec75474fb64f10674ff7fad99e43aedd6e3a90ac628d0025","source":{"kind":"arxiv","id":"2403.17804","version":1},"attestation_state":"computed","paper":{"title":"Improving Text-to-Image Consistency via Automatic Prompt Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Adina Williams, Adriana Romero-Soriano, Aishwarya Agrawal, Candace Ross, Jack Urbanek, Melissa Hall, Michal Drozdzal, Oscar Ma\\~nas, Pietro Astolfi","submitted_at":"2024-03-26T15:42:01Z","abstract_excerpt":"Impressive advances in text-to-image (T2I) generative models have yielded a plethora of high performing models which are able to generate aesthetically appealing, photorealistic images. Despite the progress, these models still struggle to produce images that are consistent with the input prompt, oftentimes failing to capture object quantities, relations and attributes properly. Existing solutions to improve prompt-image consistency suffer from the following challenges: (1) they oftentimes require model fine-tuning, (2) they only focus on nearby prompt samples, and (3) they are affected by unfa"},"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":"2403.17804","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-26T15:42:01Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c1a0c00b82505ebfd53bf48d260f288a19fc9f634ef634878b8eacd97c37ef5b","abstract_canon_sha256":"844318e6128697bc761e6261bf0494ea39fbceef2fc57c8cdafc3acedab82031"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:00:55.563204Z","signature_b64":"/CuFNtTu1llQQvFXwn1XyIJUSCAfCa3LraHkEJbhN4RF46G4NNAIbP35VQ44TLtgscPYz9R5kR/GUmQSCDmQCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a0e70072661d1690ec75474fb64f10674ff7fad99e43aedd6e3a90ac628d0025","last_reissued_at":"2026-07-05T08:00:55.562733Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:00:55.562733Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Text-to-Image Consistency via Automatic Prompt Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Adina Williams, Adriana Romero-Soriano, Aishwarya Agrawal, Candace Ross, Jack Urbanek, Melissa Hall, Michal Drozdzal, Oscar Ma\\~nas, Pietro Astolfi","submitted_at":"2024-03-26T15:42:01Z","abstract_excerpt":"Impressive advances in text-to-image (T2I) generative models have yielded a plethora of high performing models which are able to generate aesthetically appealing, photorealistic images. Despite the progress, these models still struggle to produce images that are consistent with the input prompt, oftentimes failing to capture object quantities, relations and attributes properly. Existing solutions to improve prompt-image consistency suffer from the following challenges: (1) they oftentimes require model fine-tuning, (2) they only focus on nearby prompt samples, and (3) they are affected by unfa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.17804","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/2403.17804/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":"2403.17804","created_at":"2026-07-05T08:00:55.562794+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.17804v1","created_at":"2026-07-05T08:00:55.562794+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.17804","created_at":"2026-07-05T08:00:55.562794+00:00"},{"alias_kind":"pith_short_12","alias_value":"UDTQA4TGDULJ","created_at":"2026-07-05T08:00:55.562794+00:00"},{"alias_kind":"pith_short_16","alias_value":"UDTQA4TGDULJB3DV","created_at":"2026-07-05T08:00:55.562794+00:00"},{"alias_kind":"pith_short_8","alias_value":"UDTQA4TG","created_at":"2026-07-05T08:00:55.562794+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24004","citing_title":"Towards Spec Learning: Inference-Time Alignment from Preference Pairs","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13303","citing_title":"DuET: Dual Expert Trajectories for Diffusion Image Editing","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04545","citing_title":"Impostor: An Agent-Curated Benchmark for Realistic AIGC Manipulation Localization","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24004","citing_title":"Towards Spec Learning: Inference-Time Alignment from Preference Pairs","ref_index":66,"is_internal_anchor":false},{"citing_arxiv_id":"2411.15115","citing_title":"Self-Correcting Text-to-Video Generation with Misalignment Detection and Localized Refinement","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2510.20206","citing_title":"RAPO++: Cross-Stage Prompt Optimization for Text-to-Video Generation via Data Alignment and Test-Time Scaling","ref_index":84,"is_internal_anchor":false},{"citing_arxiv_id":"2601.00090","citing_title":"It's Never Too Late: Noise Optimization for Collapse Recovery in Trained Diffusion Models","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25427","citing_title":"A Systematic Post-Train Framework for Video Generation","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12617","citing_title":"SOAR: Self-Correction for Optimal Alignment and Refinement in Diffusion Models","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07427","citing_title":"Personalizing Text-to-Image Generation to Individual Taste","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17488","citing_title":"AutoVQA-G: Self-Improving Agentic Framework for Automated Visual Question Answering and Grounding Annotation","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5","json":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5.json","graph_json":"https://pith.science/api/pith-number/UDTQA4TGDULJB3DVI5H3MTYQM5/graph.json","events_json":"https://pith.science/api/pith-number/UDTQA4TGDULJB3DVI5H3MTYQM5/events.json","paper":"https://pith.science/paper/UDTQA4TG"},"agent_actions":{"view_html":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5","download_json":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5.json","view_paper":"https://pith.science/paper/UDTQA4TG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.17804&json=true","fetch_graph":"https://pith.science/api/pith-number/UDTQA4TGDULJB3DVI5H3MTYQM5/graph.json","fetch_events":"https://pith.science/api/pith-number/UDTQA4TGDULJB3DVI5H3MTYQM5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5/action/storage_attestation","attest_author":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5/action/author_attestation","sign_citation":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5/action/citation_signature","submit_replication":"https://pith.science/pith/UDTQA4TGDULJB3DVI5H3MTYQM5/action/replication_record"}},"created_at":"2026-07-05T08:00:55.562794+00:00","updated_at":"2026-07-05T08:00:55.562794+00:00"}