{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:FBQ6YNIEF6IWAJVVY5GNUXAM6C","short_pith_number":"pith:FBQ6YNIE","schema_version":"1.0","canonical_sha256":"2861ec35042f916026b5c74cda5c0cf0aa2bcf785cca04a115dc55bd311446c5","source":{"kind":"arxiv","id":"2101.04702","version":5},"attestation_state":"computed","paper":{"title":"Cross-Modal Contrastive Learning for Text-to-Image Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Han Zhang, Honglak Lee, Jason Baldridge, Jing Yu Koh, Yinfei Yang","submitted_at":"2021-01-12T19:16:51Z","abstract_excerpt":"The output of text-to-image synthesis systems should be coherent, clear, photo-realistic scenes with high semantic fidelity to their conditioned text descriptions. Our Cross-Modal Contrastive Generative Adversarial Network (XMC-GAN) addresses this challenge by maximizing the mutual information between image and text. It does this via multiple contrastive losses which capture inter-modality and intra-modality correspondences. XMC-GAN uses an attentional self-modulation generator, which enforces strong text-image correspondence, and a contrastive discriminator, which acts as a critic as well as "},"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":"2101.04702","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-01-12T19:16:51Z","cross_cats_sorted":[],"title_canon_sha256":"eb28c0ed9d6003bb4fe5e689a10ae6fdd3f7b385906053fa33950508b53be8d2","abstract_canon_sha256":"e30ce8bb8797516d62a17732c1a6d014e76f9719787c7b6902eeddbfb7d5db4a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:14:36.532745Z","signature_b64":"tP3/TTPLq0NpiIpiBezLkWSfvRVB0SJ+Nh2y3cqASA/fOJo6IwlNGSBUML6AvA6XTHlpXJfHSDpUNzE1S1J+Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2861ec35042f916026b5c74cda5c0cf0aa2bcf785cca04a115dc55bd311446c5","last_reissued_at":"2026-07-05T04:14:36.532263Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:14:36.532263Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Cross-Modal Contrastive Learning for Text-to-Image Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Han Zhang, Honglak Lee, Jason Baldridge, Jing Yu Koh, Yinfei Yang","submitted_at":"2021-01-12T19:16:51Z","abstract_excerpt":"The output of text-to-image synthesis systems should be coherent, clear, photo-realistic scenes with high semantic fidelity to their conditioned text descriptions. Our Cross-Modal Contrastive Generative Adversarial Network (XMC-GAN) addresses this challenge by maximizing the mutual information between image and text. It does this via multiple contrastive losses which capture inter-modality and intra-modality correspondences. XMC-GAN uses an attentional self-modulation generator, which enforces strong text-image correspondence, and a contrastive discriminator, which acts as a critic as well as "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2101.04702","kind":"arxiv","version":5},"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/2101.04702/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":"2101.04702","created_at":"2026-07-05T04:14:36.532323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2101.04702v5","created_at":"2026-07-05T04:14:36.532323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2101.04702","created_at":"2026-07-05T04:14:36.532323+00:00"},{"alias_kind":"pith_short_12","alias_value":"FBQ6YNIEF6IW","created_at":"2026-07-05T04:14:36.532323+00:00"},{"alias_kind":"pith_short_16","alias_value":"FBQ6YNIEF6IWAJVV","created_at":"2026-07-05T04:14:36.532323+00:00"},{"alias_kind":"pith_short_8","alias_value":"FBQ6YNIE","created_at":"2026-07-05T04:14:36.532323+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2112.10741","citing_title":"GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2204.06125","citing_title":"Hierarchical Text-Conditional Image Generation with CLIP Latents","ref_index":59,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C","json":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C.json","graph_json":"https://pith.science/api/pith-number/FBQ6YNIEF6IWAJVVY5GNUXAM6C/graph.json","events_json":"https://pith.science/api/pith-number/FBQ6YNIEF6IWAJVVY5GNUXAM6C/events.json","paper":"https://pith.science/paper/FBQ6YNIE"},"agent_actions":{"view_html":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C","download_json":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C.json","view_paper":"https://pith.science/paper/FBQ6YNIE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2101.04702&json=true","fetch_graph":"https://pith.science/api/pith-number/FBQ6YNIEF6IWAJVVY5GNUXAM6C/graph.json","fetch_events":"https://pith.science/api/pith-number/FBQ6YNIEF6IWAJVVY5GNUXAM6C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C/action/storage_attestation","attest_author":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C/action/author_attestation","sign_citation":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C/action/citation_signature","submit_replication":"https://pith.science/pith/FBQ6YNIEF6IWAJVVY5GNUXAM6C/action/replication_record"}},"created_at":"2026-07-05T04:14:36.532323+00:00","updated_at":"2026-07-05T04:14:36.532323+00:00"}