{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:XABEDBCQAJN74CUFXML5RFOB5X","short_pith_number":"pith:XABEDBCQ","schema_version":"1.0","canonical_sha256":"b802418450025bfe0a85bb17d895c1edeaa09baf9e09a3bfc7788abf365f9ff9","source":{"kind":"arxiv","id":"2003.08936","version":4},"attestation_state":"computed","paper":{"title":"GAN Compression: Efficient Architectures for Interactive Conditional GANs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.IV"],"primary_cat":"cs.CV","authors_text":"Ji Lin, Jun-Yan Zhu, Muyang Li, Song Han, Yaoyao Ding, Zhijian Liu","submitted_at":"2020-03-19T17:59:05Z","abstract_excerpt":"Conditional Generative Adversarial Networks (cGANs) have enabled controllable image synthesis for many vision and graphics applications. However, recent cGANs are 1-2 orders of magnitude more compute-intensive than modern recognition CNNs. For example, GauGAN consumes 281G MACs per image, compared to 0.44G MACs for MobileNet-v3, making it difficult for interactive deployment. In this work, we propose a general-purpose compression framework for reducing the inference time and model size of the generator in cGANs. Directly applying existing compression methods yields poor performance due to the "},"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":"2003.08936","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-03-19T17:59:05Z","cross_cats_sorted":["eess.IV"],"title_canon_sha256":"7482af38b306569ee14c6c169dc27b67ec258e6b728874999e2e04ad3b788301","abstract_canon_sha256":"812a6a0f054cef1c54d7e7bfd00b9fe3f996322ee76d3e7ae78d6bbca50f63eb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:30:49.715645Z","signature_b64":"I0W7bF5N1xTnTP4A4qNz2u5PdGnSRSEFEkT56/G/ZEOqszHGFCg0C3qNgKk/Tj9u169ABAtYtMsbfQ8ZFW6yCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b802418450025bfe0a85bb17d895c1edeaa09baf9e09a3bfc7788abf365f9ff9","last_reissued_at":"2026-07-05T03:30:49.715266Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:30:49.715266Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GAN Compression: Efficient Architectures for Interactive Conditional GANs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.IV"],"primary_cat":"cs.CV","authors_text":"Ji Lin, Jun-Yan Zhu, Muyang Li, Song Han, Yaoyao Ding, Zhijian Liu","submitted_at":"2020-03-19T17:59:05Z","abstract_excerpt":"Conditional Generative Adversarial Networks (cGANs) have enabled controllable image synthesis for many vision and graphics applications. However, recent cGANs are 1-2 orders of magnitude more compute-intensive than modern recognition CNNs. For example, GauGAN consumes 281G MACs per image, compared to 0.44G MACs for MobileNet-v3, making it difficult for interactive deployment. In this work, we propose a general-purpose compression framework for reducing the inference time and model size of the generator in cGANs. Directly applying existing compression methods yields poor performance due to the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.08936","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/2003.08936/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":"2003.08936","created_at":"2026-07-05T03:30:49.715323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.08936v4","created_at":"2026-07-05T03:30:49.715323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.08936","created_at":"2026-07-05T03:30:49.715323+00:00"},{"alias_kind":"pith_short_12","alias_value":"XABEDBCQAJN7","created_at":"2026-07-05T03:30:49.715323+00:00"},{"alias_kind":"pith_short_16","alias_value":"XABEDBCQAJN74CUF","created_at":"2026-07-05T03:30:49.715323+00:00"},{"alias_kind":"pith_short_8","alias_value":"XABEDBCQ","created_at":"2026-07-05T03:30:49.715323+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X","json":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X.json","graph_json":"https://pith.science/api/pith-number/XABEDBCQAJN74CUFXML5RFOB5X/graph.json","events_json":"https://pith.science/api/pith-number/XABEDBCQAJN74CUFXML5RFOB5X/events.json","paper":"https://pith.science/paper/XABEDBCQ"},"agent_actions":{"view_html":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X","download_json":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X.json","view_paper":"https://pith.science/paper/XABEDBCQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.08936&json=true","fetch_graph":"https://pith.science/api/pith-number/XABEDBCQAJN74CUFXML5RFOB5X/graph.json","fetch_events":"https://pith.science/api/pith-number/XABEDBCQAJN74CUFXML5RFOB5X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X/action/storage_attestation","attest_author":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X/action/author_attestation","sign_citation":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X/action/citation_signature","submit_replication":"https://pith.science/pith/XABEDBCQAJN74CUFXML5RFOB5X/action/replication_record"}},"created_at":"2026-07-05T03:30:49.715323+00:00","updated_at":"2026-07-05T03:30:49.715323+00:00"}