{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:QD23NFP3XYYQFEP7SOSAPF3ALI","short_pith_number":"pith:QD23NFP3","schema_version":"1.0","canonical_sha256":"80f5b695fbbe310291ff93a40797605a05aecbe2df14089b89c94cf548ed216a","source":{"kind":"arxiv","id":"2006.10738","version":4},"attestation_state":"computed","paper":{"title":"Differentiable Augmentation for Data-Efficient GAN Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Ji Lin, Jun-Yan Zhu, Shengyu Zhao, Song Han, Zhijian Liu","submitted_at":"2020-06-18T17:59:01Z","abstract_excerpt":"The performance of generative adversarial networks (GANs) heavily deteriorates given a limited amount of training data. This is mainly because the discriminator is memorizing the exact training set. To combat it, we propose Differentiable Augmentation (DiffAugment), a simple method that improves the data efficiency of GANs by imposing various types of differentiable augmentations on both real and fake samples. Previous attempts to directly augment the training data manipulate the distribution of real images, yielding little benefit; DiffAugment enables us to adopt the differentiable augmentati"},"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":"2006.10738","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-06-18T17:59:01Z","cross_cats_sorted":["cs.GR","cs.LG"],"title_canon_sha256":"6b649b91f5092f201aba719f2d5c04d60b7258cac73dce93bf4a57d6155557b4","abstract_canon_sha256":"18b838f1c8127ccb14fb3f5d2158350cd10398455ef62de6b5f2356044d979ed"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:57:17.497016Z","signature_b64":"o3WTZ20ITcBixAg6fzutJzq3XwCrGBCSYUhN892vF89kmHqIFJhHYBUrb2nWx9Ld8i9P6Kj+QXCrJzsedOInCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80f5b695fbbe310291ff93a40797605a05aecbe2df14089b89c94cf548ed216a","last_reissued_at":"2026-07-05T01:57:17.496488Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:57:17.496488Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Differentiable Augmentation for Data-Efficient GAN Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Ji Lin, Jun-Yan Zhu, Shengyu Zhao, Song Han, Zhijian Liu","submitted_at":"2020-06-18T17:59:01Z","abstract_excerpt":"The performance of generative adversarial networks (GANs) heavily deteriorates given a limited amount of training data. This is mainly because the discriminator is memorizing the exact training set. To combat it, we propose Differentiable Augmentation (DiffAugment), a simple method that improves the data efficiency of GANs by imposing various types of differentiable augmentations on both real and fake samples. Previous attempts to directly augment the training data manipulate the distribution of real images, yielding little benefit; DiffAugment enables us to adopt the differentiable augmentati"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.10738","kind":"arxiv","version":4},"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/2006.10738/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":"2006.10738","created_at":"2026-07-05T01:57:17.496578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.10738v4","created_at":"2026-07-05T01:57:17.496578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.10738","created_at":"2026-07-05T01:57:17.496578+00:00"},{"alias_kind":"pith_short_12","alias_value":"QD23NFP3XYYQ","created_at":"2026-07-05T01:57:17.496578+00:00"},{"alias_kind":"pith_short_16","alias_value":"QD23NFP3XYYQFEP7","created_at":"2026-07-05T01:57:17.496578+00:00"},{"alias_kind":"pith_short_8","alias_value":"QD23NFP3","created_at":"2026-07-05T01:57:17.496578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.06678","citing_title":"A Wasserstein GAN-based climate scenario generator for risk management and insurance: the case of soil subsidence","ref_index":69,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI","json":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI.json","graph_json":"https://pith.science/api/pith-number/QD23NFP3XYYQFEP7SOSAPF3ALI/graph.json","events_json":"https://pith.science/api/pith-number/QD23NFP3XYYQFEP7SOSAPF3ALI/events.json","paper":"https://pith.science/paper/QD23NFP3"},"agent_actions":{"view_html":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI","download_json":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI.json","view_paper":"https://pith.science/paper/QD23NFP3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.10738&json=true","fetch_graph":"https://pith.science/api/pith-number/QD23NFP3XYYQFEP7SOSAPF3ALI/graph.json","fetch_events":"https://pith.science/api/pith-number/QD23NFP3XYYQFEP7SOSAPF3ALI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI/action/storage_attestation","attest_author":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI/action/author_attestation","sign_citation":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI/action/citation_signature","submit_replication":"https://pith.science/pith/QD23NFP3XYYQFEP7SOSAPF3ALI/action/replication_record"}},"created_at":"2026-07-05T01:57:17.496578+00:00","updated_at":"2026-07-05T01:57:17.496578+00:00"}