{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:6LHAXHGBV4CT6YU2MCJNKM54ES","short_pith_number":"pith:6LHAXHGB","schema_version":"1.0","canonical_sha256":"f2ce0b9cc1af053f629a6092d533bc24bb4cab8a9ec8bf8eae6b9f915c277735","source":{"kind":"arxiv","id":"2106.00563","version":3},"attestation_state":"computed","paper":{"title":"IID-GAN: an IID Sampling Perspective for Regularizing Mode Collapse","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Junchi Yan, Liangliang Shi, Yang Li","submitted_at":"2021-06-01T15:20:34Z","abstract_excerpt":"Despite its success, generative adversarial networks (GANs) still suffer from mode collapse, i.e., the generator can only map latent variables to a partial set of modes in the target distribution. In this paper, we analyze and seek to regularize this issue with an independent and identically distributed (IID) sampling perspective and emphasize that holding the IID property referring to the target distribution for generation can naturally avoid mode collapse. This is based on the basic IID assumption for real data in machine learning. However, though the source samples {z} obey IID, the generat"},"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":"2106.00563","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-06-01T15:20:34Z","cross_cats_sorted":[],"title_canon_sha256":"0bb538c911cb1b8c8add2a797bb769b06fdede073beb2fd798c5ea7841b2bea1","abstract_canon_sha256":"115f833e8f822e0a416bdb0482a73dabd53af571e4a81314e6bdb85d7c001f77"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:21:44.863355Z","signature_b64":"+qwVonDHbLhM8w1DOYOGF9HCl1AAEy6Ppl++Tgrh1m02ykD+NNNKEuPjAYkHMeEDs8RgQa1BjEDxiTIO/RnsDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f2ce0b9cc1af053f629a6092d533bc24bb4cab8a9ec8bf8eae6b9f915c277735","last_reissued_at":"2026-07-05T06:21:44.862785Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:21:44.862785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"IID-GAN: an IID Sampling Perspective for Regularizing Mode Collapse","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Junchi Yan, Liangliang Shi, Yang Li","submitted_at":"2021-06-01T15:20:34Z","abstract_excerpt":"Despite its success, generative adversarial networks (GANs) still suffer from mode collapse, i.e., the generator can only map latent variables to a partial set of modes in the target distribution. In this paper, we analyze and seek to regularize this issue with an independent and identically distributed (IID) sampling perspective and emphasize that holding the IID property referring to the target distribution for generation can naturally avoid mode collapse. This is based on the basic IID assumption for real data in machine learning. However, though the source samples {z} obey IID, the generat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.00563","kind":"arxiv","version":3},"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/2106.00563/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":"2106.00563","created_at":"2026-07-05T06:21:44.862850+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.00563v3","created_at":"2026-07-05T06:21:44.862850+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.00563","created_at":"2026-07-05T06:21:44.862850+00:00"},{"alias_kind":"pith_short_12","alias_value":"6LHAXHGBV4CT","created_at":"2026-07-05T06:21:44.862850+00:00"},{"alias_kind":"pith_short_16","alias_value":"6LHAXHGBV4CT6YU2","created_at":"2026-07-05T06:21:44.862850+00:00"},{"alias_kind":"pith_short_8","alias_value":"6LHAXHGB","created_at":"2026-07-05T06:21:44.862850+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.11236","citing_title":"A New Formulation of Lipschitz Constrained With Functional Gradient Learning for GANs","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES","json":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES.json","graph_json":"https://pith.science/api/pith-number/6LHAXHGBV4CT6YU2MCJNKM54ES/graph.json","events_json":"https://pith.science/api/pith-number/6LHAXHGBV4CT6YU2MCJNKM54ES/events.json","paper":"https://pith.science/paper/6LHAXHGB"},"agent_actions":{"view_html":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES","download_json":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES.json","view_paper":"https://pith.science/paper/6LHAXHGB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.00563&json=true","fetch_graph":"https://pith.science/api/pith-number/6LHAXHGBV4CT6YU2MCJNKM54ES/graph.json","fetch_events":"https://pith.science/api/pith-number/6LHAXHGBV4CT6YU2MCJNKM54ES/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES/action/storage_attestation","attest_author":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES/action/author_attestation","sign_citation":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES/action/citation_signature","submit_replication":"https://pith.science/pith/6LHAXHGBV4CT6YU2MCJNKM54ES/action/replication_record"}},"created_at":"2026-07-05T06:21:44.862850+00:00","updated_at":"2026-07-05T06:21:44.862850+00:00"}