{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:XR7ZCHU7NJYIB5QTNHFSMXCTF3","short_pith_number":"pith:XR7ZCHU7","schema_version":"1.0","canonical_sha256":"bc7f911e9f6a7080f61369cb265c532eceb220f0038ff2204061a95740d63b19","source":{"kind":"arxiv","id":"2305.13655","version":3},"attestation_state":"computed","paper":{"title":"LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adam Yala, Boyi Li, Long Lian, Trevor Darrell","submitted_at":"2023-05-23T03:59:06Z","abstract_excerpt":"Recent advancements in text-to-image diffusion models have yielded impressive results in generating realistic and diverse images. However, these models still struggle with complex prompts, such as those that involve numeracy and spatial reasoning. This work proposes to enhance prompt understanding capabilities in diffusion models. Our method leverages a pretrained large language model (LLM) for grounded generation in a novel two-stage process. In the first stage, the LLM generates a scene layout that comprises captioned bounding boxes from a given prompt describing the desired image. In the se"},"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":"2305.13655","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-05-23T03:59:06Z","cross_cats_sorted":[],"title_canon_sha256":"68552a17253802c6156292939388ab77b27675e3e2c0c721408a6d7db74de4aa","abstract_canon_sha256":"d06de464760dac2b53953b10a3b5deec725c0a5d3514651c193d3baaa0844df2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:51:28.392105Z","signature_b64":"yst0BS06z+FQiFx8BpDfFBX6RilqTB5QJYqGs7Co10CUx8caYO5VoWFUl4C5NwctwDNdyMqIiQRajhLdiOsCCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc7f911e9f6a7080f61369cb265c532eceb220f0038ff2204061a95740d63b19","last_reissued_at":"2026-07-05T07:51:28.391570Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:51:28.391570Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adam Yala, Boyi Li, Long Lian, Trevor Darrell","submitted_at":"2023-05-23T03:59:06Z","abstract_excerpt":"Recent advancements in text-to-image diffusion models have yielded impressive results in generating realistic and diverse images. However, these models still struggle with complex prompts, such as those that involve numeracy and spatial reasoning. This work proposes to enhance prompt understanding capabilities in diffusion models. Our method leverages a pretrained large language model (LLM) for grounded generation in a novel two-stage process. In the first stage, the LLM generates a scene layout that comprises captioned bounding boxes from a given prompt describing the desired image. In the se"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.13655","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/2305.13655/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":"2305.13655","created_at":"2026-07-05T07:51:28.391627+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.13655v3","created_at":"2026-07-05T07:51:28.391627+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.13655","created_at":"2026-07-05T07:51:28.391627+00:00"},{"alias_kind":"pith_short_12","alias_value":"XR7ZCHU7NJYI","created_at":"2026-07-05T07:51:28.391627+00:00"},{"alias_kind":"pith_short_16","alias_value":"XR7ZCHU7NJYIB5QT","created_at":"2026-07-05T07:51:28.391627+00:00"},{"alias_kind":"pith_short_8","alias_value":"XR7ZCHU7","created_at":"2026-07-05T07:51:28.391627+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24849","citing_title":"IV-CoT: Implicit Visual Chain-of-Thought for Structure-Aware Text-to-Image Generation","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13247","citing_title":"EPIG: Emotion-Based Prompting for Personalised Image Generation","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10653","citing_title":"STEDiff: Strengthening Text Embedding for Text-to-Image Alignment in Diffusion Model","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28615","citing_title":"Compositional Text-to-Image Generation Via Region-aware Bimodal Direct Preference Optimization","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2404.10141","citing_title":"ANCHOR: LLM-driven Subject Conditioning for Text-to-Image Synthesis","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2411.19182","citing_title":"SOWing Information: Cultivating Contextual Coherence with MLLMs in Image Generation","ref_index":54,"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":21,"is_internal_anchor":false},{"citing_arxiv_id":"2404.16994","citing_title":"PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2405.07987","citing_title":"The Platonic Representation Hypothesis","ref_index":271,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11722","citing_title":"EPIC: Efficient Predicate-Guided Inference-Time Control for Compositional Text-to-Image Generation","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25477","citing_title":"DDA-Thinker: Decoupled Dual-Atomic Reinforcement Learning for Reasoning-Driven Image Editing","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2403.05135","citing_title":"ELLA: Equip Diffusion Models with LLM for Enhanced Semantic Alignment","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14062","citing_title":"OneHOI: Unifying Human-Object Interaction Generation and Editing","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15917","citing_title":"Making Image Editing Easier via Adaptive Task Reformulation with Agentic Executions","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17749","citing_title":"Ego-InBetween: Generating Object State Transitions in Ego-Centric Videos","ref_index":25,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3","json":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3.json","graph_json":"https://pith.science/api/pith-number/XR7ZCHU7NJYIB5QTNHFSMXCTF3/graph.json","events_json":"https://pith.science/api/pith-number/XR7ZCHU7NJYIB5QTNHFSMXCTF3/events.json","paper":"https://pith.science/paper/XR7ZCHU7"},"agent_actions":{"view_html":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3","download_json":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3.json","view_paper":"https://pith.science/paper/XR7ZCHU7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.13655&json=true","fetch_graph":"https://pith.science/api/pith-number/XR7ZCHU7NJYIB5QTNHFSMXCTF3/graph.json","fetch_events":"https://pith.science/api/pith-number/XR7ZCHU7NJYIB5QTNHFSMXCTF3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3/action/storage_attestation","attest_author":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3/action/author_attestation","sign_citation":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3/action/citation_signature","submit_replication":"https://pith.science/pith/XR7ZCHU7NJYIB5QTNHFSMXCTF3/action/replication_record"}},"created_at":"2026-07-05T07:51:28.391627+00:00","updated_at":"2026-07-05T07:51:28.391627+00:00"}