{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5IEUCGG7JZJXJO6BXBECD7MEHV","short_pith_number":"pith:5IEUCGG7","schema_version":"1.0","canonical_sha256":"ea094118df4e5374bbc1b84821fd843d7aeddbbec0ae82919e91b7f1f55165ff","source":{"kind":"arxiv","id":"2403.18493","version":1},"attestation_state":"computed","paper":{"title":"VersaT2I: Improving Text-to-Image Models with Versatile Reward","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gaoang Wang, Hsiang-Wei Huang, Jenq-Neng Hwang, Jianshu Guo, Jiawei Zhang, Jie Deng, Tian Ye, Wenhao Chai, Yichen Xu","submitted_at":"2024-03-27T12:08:41Z","abstract_excerpt":"Recent text-to-image (T2I) models have benefited from large-scale and high-quality data, demonstrating impressive performance. However, these T2I models still struggle to produce images that are aesthetically pleasing, geometrically accurate, faithful to text, and of good low-level quality. We present VersaT2I, a versatile training framework that can boost the performance with multiple rewards of any T2I model. We decompose the quality of the image into several aspects such as aesthetics, text-image alignment, geometry, low-level quality, etc. Then, for every quality aspect, we select high-qua"},"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":"2403.18493","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-27T12:08:41Z","cross_cats_sorted":[],"title_canon_sha256":"5c15f676b6cfa187e71aa819999a1a0d7f6a1ef317409aaa98ed92f679066f64","abstract_canon_sha256":"5eb5e1b961bb546b14831059ad26d4b48b2d73ee0ec24a5343eb36e3e8f5e3bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:01:23.528437Z","signature_b64":"FaJnBddFiKtLPlWoEsljSwLZV2OKbedtcSpzVUz4cgM/o92jKrWJM4OZ8u5If6fBwKUjWQfUqgYl0Unpif52Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ea094118df4e5374bbc1b84821fd843d7aeddbbec0ae82919e91b7f1f55165ff","last_reissued_at":"2026-07-05T08:01:23.527988Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:01:23.527988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VersaT2I: Improving Text-to-Image Models with Versatile Reward","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Gaoang Wang, Hsiang-Wei Huang, Jenq-Neng Hwang, Jianshu Guo, Jiawei Zhang, Jie Deng, Tian Ye, Wenhao Chai, Yichen Xu","submitted_at":"2024-03-27T12:08:41Z","abstract_excerpt":"Recent text-to-image (T2I) models have benefited from large-scale and high-quality data, demonstrating impressive performance. However, these T2I models still struggle to produce images that are aesthetically pleasing, geometrically accurate, faithful to text, and of good low-level quality. We present VersaT2I, a versatile training framework that can boost the performance with multiple rewards of any T2I model. We decompose the quality of the image into several aspects such as aesthetics, text-image alignment, geometry, low-level quality, etc. Then, for every quality aspect, we select high-qua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.18493","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/2403.18493/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":"2403.18493","created_at":"2026-07-05T08:01:23.528043+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.18493v1","created_at":"2026-07-05T08:01:23.528043+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.18493","created_at":"2026-07-05T08:01:23.528043+00:00"},{"alias_kind":"pith_short_12","alias_value":"5IEUCGG7JZJX","created_at":"2026-07-05T08:01:23.528043+00:00"},{"alias_kind":"pith_short_16","alias_value":"5IEUCGG7JZJXJO6B","created_at":"2026-07-05T08:01:23.528043+00:00"},{"alias_kind":"pith_short_8","alias_value":"5IEUCGG7","created_at":"2026-07-05T08:01:23.528043+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":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13305","citing_title":"Bias at the End of the Score","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV","json":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV.json","graph_json":"https://pith.science/api/pith-number/5IEUCGG7JZJXJO6BXBECD7MEHV/graph.json","events_json":"https://pith.science/api/pith-number/5IEUCGG7JZJXJO6BXBECD7MEHV/events.json","paper":"https://pith.science/paper/5IEUCGG7"},"agent_actions":{"view_html":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV","download_json":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV.json","view_paper":"https://pith.science/paper/5IEUCGG7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.18493&json=true","fetch_graph":"https://pith.science/api/pith-number/5IEUCGG7JZJXJO6BXBECD7MEHV/graph.json","fetch_events":"https://pith.science/api/pith-number/5IEUCGG7JZJXJO6BXBECD7MEHV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV/action/storage_attestation","attest_author":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV/action/author_attestation","sign_citation":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV/action/citation_signature","submit_replication":"https://pith.science/pith/5IEUCGG7JZJXJO6BXBECD7MEHV/action/replication_record"}},"created_at":"2026-07-05T08:01:23.528043+00:00","updated_at":"2026-07-05T08:01:23.528043+00:00"}