{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5TYEHFC7DNR5X3POW4C52M2T3D","short_pith_number":"pith:5TYEHFC7","schema_version":"1.0","canonical_sha256":"ecf043945f1b63dbedeeb705dd3353d8cf5c94223dd2b52ee6fcc3d33174404e","source":{"kind":"arxiv","id":"2506.01493","version":1},"attestation_state":"computed","paper":{"title":"Efficiency without Compromise: CLIP-aided Text-to-Image GANs with Increased Diversity","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Takashi Shibuya, Yuhta Takida, Yuki Mitsufuji, Yuya Kobayashi","submitted_at":"2025-06-02T09:54:41Z","abstract_excerpt":"Recently, Generative Adversarial Networks (GANs) have been successfully scaled to billion-scale large text-to-image datasets. However, training such models entails a high training cost, limiting some applications and research usage. To reduce the cost, one promising direction is the incorporation of pre-trained models. The existing method of utilizing pre-trained models for a generator significantly reduced the training cost compared with the other large-scale GANs, but we found the model loses the diversity of generation for a given prompt by a large margin. To build an efficient and high-fid"},"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":"2506.01493","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-02T09:54:41Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b979b34f9c59d819d6006ae15ad4bb4ec8f00aa3225b4172b7a8e1d5b8a56d91","abstract_canon_sha256":"cb76f4ce8e902952bca581cf77c32acae72f2aef78038b1a640b64b7d0287837"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:14.609004Z","signature_b64":"vF5PTw7xmAm2SiaKNhx8K17PC5ruPHhgbdoVZgGikuYiyJxfxnp5ek9bF5Xo+qK2yYkjj5ZVvzXrSqCyP47nAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ecf043945f1b63dbedeeb705dd3353d8cf5c94223dd2b52ee6fcc3d33174404e","last_reissued_at":"2026-07-05T11:14:14.608532Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:14.608532Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficiency without Compromise: CLIP-aided Text-to-Image GANs with Increased Diversity","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Takashi Shibuya, Yuhta Takida, Yuki Mitsufuji, Yuya Kobayashi","submitted_at":"2025-06-02T09:54:41Z","abstract_excerpt":"Recently, Generative Adversarial Networks (GANs) have been successfully scaled to billion-scale large text-to-image datasets. However, training such models entails a high training cost, limiting some applications and research usage. To reduce the cost, one promising direction is the incorporation of pre-trained models. The existing method of utilizing pre-trained models for a generator significantly reduced the training cost compared with the other large-scale GANs, but we found the model loses the diversity of generation for a given prompt by a large margin. To build an efficient and high-fid"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01493","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/2506.01493/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":"2506.01493","created_at":"2026-07-05T11:14:14.608590+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.01493v1","created_at":"2026-07-05T11:14:14.608590+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01493","created_at":"2026-07-05T11:14:14.608590+00:00"},{"alias_kind":"pith_short_12","alias_value":"5TYEHFC7DNR5","created_at":"2026-07-05T11:14:14.608590+00:00"},{"alias_kind":"pith_short_16","alias_value":"5TYEHFC7DNR5X3PO","created_at":"2026-07-05T11:14:14.608590+00:00"},{"alias_kind":"pith_short_8","alias_value":"5TYEHFC7","created_at":"2026-07-05T11:14:14.608590+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2603.02667","citing_title":"Unifying Contrastive and Generative Objectives for Visual Understanding and Text-to-Image Generation","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D","json":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D.json","graph_json":"https://pith.science/api/pith-number/5TYEHFC7DNR5X3POW4C52M2T3D/graph.json","events_json":"https://pith.science/api/pith-number/5TYEHFC7DNR5X3POW4C52M2T3D/events.json","paper":"https://pith.science/paper/5TYEHFC7"},"agent_actions":{"view_html":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D","download_json":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D.json","view_paper":"https://pith.science/paper/5TYEHFC7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.01493&json=true","fetch_graph":"https://pith.science/api/pith-number/5TYEHFC7DNR5X3POW4C52M2T3D/graph.json","fetch_events":"https://pith.science/api/pith-number/5TYEHFC7DNR5X3POW4C52M2T3D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D/action/storage_attestation","attest_author":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D/action/author_attestation","sign_citation":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D/action/citation_signature","submit_replication":"https://pith.science/pith/5TYEHFC7DNR5X3POW4C52M2T3D/action/replication_record"}},"created_at":"2026-07-05T11:14:14.608590+00:00","updated_at":"2026-07-05T11:14:14.608590+00:00"}