{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:ZHW2WLZD77YOKVYP3CF6CWHYC7","short_pith_number":"pith:ZHW2WLZD","schema_version":"1.0","canonical_sha256":"c9edab2f23fff0e5570fd88be158f817caae35564178139c639dadb99b0cfbb1","source":{"kind":"arxiv","id":"1807.03039","version":2},"attestation_state":"computed","paper":{"title":"Glow: Generative Flow with Invertible 1x1 Convolutions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Diederik P. Kingma, Prafulla Dhariwal","submitted_at":"2018-07-09T10:57:26Z","abstract_excerpt":"Flow-based generative models (Dinh et al., 2014) are conceptually attractive due to tractability of the exact log-likelihood, tractability of exact latent-variable inference, and parallelizability of both training and synthesis. In this paper we propose Glow, a simple type of generative flow using an invertible 1x1 convolution. Using our method we demonstrate a significant improvement in log-likelihood on standard benchmarks. Perhaps most strikingly, we demonstrate that a generative model optimized towards the plain log-likelihood objective is capable of efficient realistic-looking synthesis a"},"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":"1807.03039","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-07-09T10:57:26Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"b7266b5353117eb2c4fa54655828c430dc7558d35fc54a4919e50e67c4415e9e","abstract_canon_sha256":"2c6c94d192b013b9294e9602df3027783f5a7dd7637f9a2e2942b7e04e53a3d8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:11:10.882293Z","signature_b64":"17b+9l6uo+ZQU/tJAsdfXz4Py8lqKhvoK3Yu9AqO285MEDD5RzWw4kyWrZo9QQIhxnsHr1UtXWFrk0a7nvWFDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9edab2f23fff0e5570fd88be158f817caae35564178139c639dadb99b0cfbb1","last_reissued_at":"2026-05-18T00:11:10.881618Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:11:10.881618Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Glow: Generative Flow with Invertible 1x1 Convolutions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"stat.ML","authors_text":"Diederik P. Kingma, Prafulla Dhariwal","submitted_at":"2018-07-09T10:57:26Z","abstract_excerpt":"Flow-based generative models (Dinh et al., 2014) are conceptually attractive due to tractability of the exact log-likelihood, tractability of exact latent-variable inference, and parallelizability of both training and synthesis. In this paper we propose Glow, a simple type of generative flow using an invertible 1x1 convolution. Using our method we demonstrate a significant improvement in log-likelihood on standard benchmarks. Perhaps most strikingly, we demonstrate that a generative model optimized towards the plain log-likelihood objective is capable of efficient realistic-looking synthesis a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.03039","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1807.03039","created_at":"2026-05-18T00:11:10.881741+00:00"},{"alias_kind":"arxiv_version","alias_value":"1807.03039v2","created_at":"2026-05-18T00:11:10.881741+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.03039","created_at":"2026-05-18T00:11:10.881741+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZHW2WLZD77YO","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZHW2WLZD77YOKVYP","created_at":"2026-05-18T12:33:07.085635+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZHW2WLZD","created_at":"2026-05-18T12:33:07.085635+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"1907.02392","citing_title":"Guided Image Generation with Conditional Invertible Neural Networks","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.17511","citing_title":"A flow-matching generative model for event-by-event jet-induced hydro response in high-energy heavy-ion collisions","ref_index":104,"is_internal_anchor":true},{"citing_arxiv_id":"2602.03875","citing_title":"Reversible Deep Learning for 13C NMR in Chemoinformatics: On Structures and Spectra","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2104.10157","citing_title":"VideoGPT: Video Generation using VQ-VAE and Transformers","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06348","citing_title":"Dartmouth Stellar Evolution Emulator (DSEE) 1: Generative Stellar Evolution Model Database","ref_index":58,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7","json":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7.json","graph_json":"https://pith.science/api/pith-number/ZHW2WLZD77YOKVYP3CF6CWHYC7/graph.json","events_json":"https://pith.science/api/pith-number/ZHW2WLZD77YOKVYP3CF6CWHYC7/events.json","paper":"https://pith.science/paper/ZHW2WLZD"},"agent_actions":{"view_html":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7","download_json":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7.json","view_paper":"https://pith.science/paper/ZHW2WLZD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1807.03039&json=true","fetch_graph":"https://pith.science/api/pith-number/ZHW2WLZD77YOKVYP3CF6CWHYC7/graph.json","fetch_events":"https://pith.science/api/pith-number/ZHW2WLZD77YOKVYP3CF6CWHYC7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7/action/storage_attestation","attest_author":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7/action/author_attestation","sign_citation":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7/action/citation_signature","submit_replication":"https://pith.science/pith/ZHW2WLZD77YOKVYP3CF6CWHYC7/action/replication_record"}},"created_at":"2026-05-18T00:11:10.881741+00:00","updated_at":"2026-05-18T00:11:10.881741+00:00"}