{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:45KTSWNWVJ6JWDHF56V2F3CMLJ","short_pith_number":"pith:45KTSWNW","schema_version":"1.0","canonical_sha256":"e7553959b6aa7c9b0ce5efaba2ec4c5a4e9bed61107cc60c1231d22a42f44a92","source":{"kind":"arxiv","id":"2507.04451","version":1},"attestation_state":"computed","paper":{"title":"CoT-lized Diffusion: Let's Reinforce T2I Generation Step-by-step","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fan Wang, Li Yuan, Munan Ning, Qihui Zhang, Shuo Yang, Weihua Chen, Xianzhe Xu, Yibing Song, Yiwei Yang, Zheyuan Liu, Zhongrui Wang","submitted_at":"2025-07-06T16:17:32Z","abstract_excerpt":"Current text-to-image (T2I) generation models struggle to align spatial composition with the input text, especially in complex scenes. Even layout-based approaches yield suboptimal spatial control, as their generation process is decoupled from layout planning, making it difficult to refine the layout during synthesis. We present CoT-Diff, a framework that brings step-by-step CoT-style reasoning into T2I generation by tightly integrating Multimodal Large Language Model (MLLM)-driven 3D layout planning with the diffusion process. CoT-Diff enables layout-aware reasoning inline within a single dif"},"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":"2507.04451","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-07-06T16:17:32Z","cross_cats_sorted":[],"title_canon_sha256":"abe0af40f83ccdced80c8d402f8c1d5ce24c46e40a1c238e70159d04e15e9109","abstract_canon_sha256":"b1142d3b0a46dc468abfbea687f8033d15f1ad2af9575444a254ddf2872cbb16"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:48.273694Z","signature_b64":"oPL95yxwNAncvL0omtrK67gGoxeDilV6+Yu5svanvSvPBipSRjpSjW4j9PCKHgLJwMoQt7T0YznD1VZS68qsDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7553959b6aa7c9b0ce5efaba2ec4c5a4e9bed61107cc60c1231d22a42f44a92","last_reissued_at":"2026-07-05T11:32:48.273203Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:48.273203Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CoT-lized Diffusion: Let's Reinforce T2I Generation Step-by-step","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fan Wang, Li Yuan, Munan Ning, Qihui Zhang, Shuo Yang, Weihua Chen, Xianzhe Xu, Yibing Song, Yiwei Yang, Zheyuan Liu, Zhongrui Wang","submitted_at":"2025-07-06T16:17:32Z","abstract_excerpt":"Current text-to-image (T2I) generation models struggle to align spatial composition with the input text, especially in complex scenes. Even layout-based approaches yield suboptimal spatial control, as their generation process is decoupled from layout planning, making it difficult to refine the layout during synthesis. We present CoT-Diff, a framework that brings step-by-step CoT-style reasoning into T2I generation by tightly integrating Multimodal Large Language Model (MLLM)-driven 3D layout planning with the diffusion process. CoT-Diff enables layout-aware reasoning inline within a single dif"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.04451","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/2507.04451/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":"2507.04451","created_at":"2026-07-05T11:32:48.273260+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.04451v1","created_at":"2026-07-05T11:32:48.273260+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.04451","created_at":"2026-07-05T11:32:48.273260+00:00"},{"alias_kind":"pith_short_12","alias_value":"45KTSWNWVJ6J","created_at":"2026-07-05T11:32:48.273260+00:00"},{"alias_kind":"pith_short_16","alias_value":"45KTSWNWVJ6JWDHF","created_at":"2026-07-05T11:32:48.273260+00:00"},{"alias_kind":"pith_short_8","alias_value":"45KTSWNW","created_at":"2026-07-05T11:32:48.273260+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01803","citing_title":"OctoT2I: A Self-Evolving Agentic Text-to-Image Router","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2510.16888","citing_title":"Uniworld-V2: Reinforce Image Editing with Diffusion Negative-aware Finetuning and MLLM Implicit Feedback","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04609","citing_title":"Advancing Aesthetic Image Generation via Composition Transfer","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ","json":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ.json","graph_json":"https://pith.science/api/pith-number/45KTSWNWVJ6JWDHF56V2F3CMLJ/graph.json","events_json":"https://pith.science/api/pith-number/45KTSWNWVJ6JWDHF56V2F3CMLJ/events.json","paper":"https://pith.science/paper/45KTSWNW"},"agent_actions":{"view_html":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ","download_json":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ.json","view_paper":"https://pith.science/paper/45KTSWNW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.04451&json=true","fetch_graph":"https://pith.science/api/pith-number/45KTSWNWVJ6JWDHF56V2F3CMLJ/graph.json","fetch_events":"https://pith.science/api/pith-number/45KTSWNWVJ6JWDHF56V2F3CMLJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ/action/storage_attestation","attest_author":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ/action/author_attestation","sign_citation":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ/action/citation_signature","submit_replication":"https://pith.science/pith/45KTSWNWVJ6JWDHF56V2F3CMLJ/action/replication_record"}},"created_at":"2026-07-05T11:32:48.273260+00:00","updated_at":"2026-07-05T11:32:48.273260+00:00"}