{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FTXZOSBUV2ZSN4BPH4QHFOMXRS","short_pith_number":"pith:FTXZOSBU","schema_version":"1.0","canonical_sha256":"2cef974834aeb326f02f3f2072b9978cb84fa7a417a52f99949a46d0972ba729","source":{"kind":"arxiv","id":"2401.11708","version":3},"attestation_state":"computed","paper":{"title":"Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bin Cui, Chenlin Meng, Ling Yang, Minkai Xu, Stefano Ermon, Zhaochen Yu","submitted_at":"2024-01-22T06:16:29Z","abstract_excerpt":"Diffusion models have exhibit exceptional performance in text-to-image generation and editing. However, existing methods often face challenges when handling complex text prompts that involve multiple objects with multiple attributes and relationships. In this paper, we propose a brand new training-free text-to-image generation/editing framework, namely Recaption, Plan and Generate (RPG), harnessing the powerful chain-of-thought reasoning ability of multimodal LLMs to enhance the compositionality of text-to-image diffusion models. Our approach employs the MLLM as a global planner to decompose t"},"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":"2401.11708","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-01-22T06:16:29Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"5e9e6791d516987e9291850306e112b92d22b94f1ac123de0175e750c9ae2758","abstract_canon_sha256":"1428810ef164932898969c09dcbe04ef2e1d18a581821dbf96e640a9b3ebf470"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:02.683946Z","signature_b64":"8IWE+L+AZ8IRqkqj7WqthtfLp1cQ/Lcm7xpeXBOvpM17BKb4mC8vExB/hoiKgytpQyPjBLlrii+cqnpwztB8CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2cef974834aeb326f02f3f2072b9978cb84fa7a417a52f99949a46d0972ba729","last_reissued_at":"2026-07-05T08:27:02.683477Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:02.683477Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mastering Text-to-Image Diffusion: Recaptioning, Planning, and Generating with Multimodal LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bin Cui, Chenlin Meng, Ling Yang, Minkai Xu, Stefano Ermon, Zhaochen Yu","submitted_at":"2024-01-22T06:16:29Z","abstract_excerpt":"Diffusion models have exhibit exceptional performance in text-to-image generation and editing. However, existing methods often face challenges when handling complex text prompts that involve multiple objects with multiple attributes and relationships. In this paper, we propose a brand new training-free text-to-image generation/editing framework, namely Recaption, Plan and Generate (RPG), harnessing the powerful chain-of-thought reasoning ability of multimodal LLMs to enhance the compositionality of text-to-image diffusion models. Our approach employs the MLLM as a global planner to decompose t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.11708","kind":"arxiv","version":3},"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/2401.11708/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":"2401.11708","created_at":"2026-07-05T08:27:02.683534+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.11708v3","created_at":"2026-07-05T08:27:02.683534+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.11708","created_at":"2026-07-05T08:27:02.683534+00:00"},{"alias_kind":"pith_short_12","alias_value":"FTXZOSBUV2ZS","created_at":"2026-07-05T08:27:02.683534+00:00"},{"alias_kind":"pith_short_16","alias_value":"FTXZOSBUV2ZSN4BP","created_at":"2026-07-05T08:27:02.683534+00:00"},{"alias_kind":"pith_short_8","alias_value":"FTXZOSBU","created_at":"2026-07-05T08:27:02.683534+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.28615","citing_title":"Compositional Text-to-Image Generation Via Region-aware Bimodal Direct Preference Optimization","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2405.08748","citing_title":"Hunyuan-DiT: A Powerful Multi-Resolution Diffusion Transformer with Fine-Grained Chinese Understanding","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2403.05135","citing_title":"ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2504.17761","citing_title":"Step1X-Edit: A Practical Framework for General Image Editing","ref_index":62,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS","json":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS.json","graph_json":"https://pith.science/api/pith-number/FTXZOSBUV2ZSN4BPH4QHFOMXRS/graph.json","events_json":"https://pith.science/api/pith-number/FTXZOSBUV2ZSN4BPH4QHFOMXRS/events.json","paper":"https://pith.science/paper/FTXZOSBU"},"agent_actions":{"view_html":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS","download_json":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS.json","view_paper":"https://pith.science/paper/FTXZOSBU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.11708&json=true","fetch_graph":"https://pith.science/api/pith-number/FTXZOSBUV2ZSN4BPH4QHFOMXRS/graph.json","fetch_events":"https://pith.science/api/pith-number/FTXZOSBUV2ZSN4BPH4QHFOMXRS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS/action/storage_attestation","attest_author":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS/action/author_attestation","sign_citation":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS/action/citation_signature","submit_replication":"https://pith.science/pith/FTXZOSBUV2ZSN4BPH4QHFOMXRS/action/replication_record"}},"created_at":"2026-07-05T08:27:02.683534+00:00","updated_at":"2026-07-05T08:27:02.683534+00:00"}