{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:P4QK3JG2WZ36CWAZU7OIJ2ZRJZ","short_pith_number":"pith:P4QK3JG2","schema_version":"1.0","canonical_sha256":"7f20ada4dab677e15819a7dc84eb314e4857ca41703c8c4899d489c2fe1ac821","source":{"kind":"arxiv","id":"2404.01197","version":2},"attestation_state":"computed","paper":{"title":"Getting it Right: Improving Spatial Consistency in Text-to-Image Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Agneet Chatterjee, Chitta Baral, Dhruba Ghosh, Estelle Aflalo, Gabriela Ben Melech Stan, Hannaneh Hajishirzi, Ludwig Schmidt, Sayak Paul, Tejas Gokhale, Vasudev Lal, Yezhou Yang","submitted_at":"2024-04-01T15:55:25Z","abstract_excerpt":"One of the key shortcomings in current text-to-image (T2I) models is their inability to consistently generate images which faithfully follow the spatial relationships specified in the text prompt. In this paper, we offer a comprehensive investigation of this limitation, while also developing datasets and methods that support algorithmic solutions to improve spatial reasoning in T2I models. We find that spatial relationships are under-represented in the image descriptions found in current vision-language datasets. To alleviate this data bottleneck, we create SPRIGHT, the first spatially focused"},"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":"2404.01197","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-04-01T15:55:25Z","cross_cats_sorted":[],"title_canon_sha256":"c8588564af6fdb4fdd9af63af3e3e7a8cba40536d832d1efe8dba4f74718b02b","abstract_canon_sha256":"1f0e18feb45257c2302b1d6f4dc3bf597313b68f35c81128b97fe5efd493de5a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:52:31.311114Z","signature_b64":"tS7iyANQnYl366e/XFWU5jp1MR1qoMPCc5vQuZoHucnesF3pIponHixpk7Y3vWTpi7B/x3/D1IQ76AnsaIooBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7f20ada4dab677e15819a7dc84eb314e4857ca41703c8c4899d489c2fe1ac821","last_reissued_at":"2026-07-05T08:52:31.310650Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:52:31.310650Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Getting it Right: Improving Spatial Consistency in Text-to-Image Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Agneet Chatterjee, Chitta Baral, Dhruba Ghosh, Estelle Aflalo, Gabriela Ben Melech Stan, Hannaneh Hajishirzi, Ludwig Schmidt, Sayak Paul, Tejas Gokhale, Vasudev Lal, Yezhou Yang","submitted_at":"2024-04-01T15:55:25Z","abstract_excerpt":"One of the key shortcomings in current text-to-image (T2I) models is their inability to consistently generate images which faithfully follow the spatial relationships specified in the text prompt. In this paper, we offer a comprehensive investigation of this limitation, while also developing datasets and methods that support algorithmic solutions to improve spatial reasoning in T2I models. We find that spatial relationships are under-represented in the image descriptions found in current vision-language datasets. To alleviate this data bottleneck, we create SPRIGHT, the first spatially focused"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01197","kind":"arxiv","version":2},"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/2404.01197/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":"2404.01197","created_at":"2026-07-05T08:52:31.310708+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.01197v2","created_at":"2026-07-05T08:52:31.310708+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01197","created_at":"2026-07-05T08:52:31.310708+00:00"},{"alias_kind":"pith_short_12","alias_value":"P4QK3JG2WZ36","created_at":"2026-07-05T08:52:31.310708+00:00"},{"alias_kind":"pith_short_16","alias_value":"P4QK3JG2WZ36CWAZ","created_at":"2026-07-05T08:52:31.310708+00:00"},{"alias_kind":"pith_short_8","alias_value":"P4QK3JG2","created_at":"2026-07-05T08:52:31.310708+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.17458","citing_title":"CARINOX: Inference-time Scaling with Category-Aware Reward-based Initial Noise Optimization and Exploration","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2509.17458","citing_title":"CARINOX: Inference-time Scaling with Category-Aware Reward-based Initial Noise Optimization and Exploration","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ","json":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ.json","graph_json":"https://pith.science/api/pith-number/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/graph.json","events_json":"https://pith.science/api/pith-number/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/events.json","paper":"https://pith.science/paper/P4QK3JG2"},"agent_actions":{"view_html":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ","download_json":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ.json","view_paper":"https://pith.science/paper/P4QK3JG2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.01197&json=true","fetch_graph":"https://pith.science/api/pith-number/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/graph.json","fetch_events":"https://pith.science/api/pith-number/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/action/storage_attestation","attest_author":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/action/author_attestation","sign_citation":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/action/citation_signature","submit_replication":"https://pith.science/pith/P4QK3JG2WZ36CWAZU7OIJ2ZRJZ/action/replication_record"}},"created_at":"2026-07-05T08:52:31.310708+00:00","updated_at":"2026-07-05T08:52:31.310708+00:00"}