{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:GUKCBAIVYI64TFXM7RMPVBKI2E","short_pith_number":"pith:GUKCBAIV","schema_version":"1.0","canonical_sha256":"3514208115c23dc996ecfc58fa8548d120163ddd57dc7f77b865bce166dd1a4d","source":{"kind":"arxiv","id":"2009.05475","version":2},"attestation_state":"computed","paper":{"title":"Adversarial score matching and improved sampling for image generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexia Jolicoeur-Martineau, Ioannis Mitliagkas, R\\'emi Pich\\'e-Taillefer, R\\'emi Tachet des Combes","submitted_at":"2020-09-11T14:49:53Z","abstract_excerpt":"Denoising Score Matching with Annealed Langevin Sampling (DSM-ALS) has recently found success in generative modeling. The approach works by first training a neural network to estimate the score of a distribution, and then using Langevin dynamics to sample from the data distribution assumed by the score network. Despite the convincing visual quality of samples, this method appears to perform worse than Generative Adversarial Networks (GANs) under the Fr\\'echet Inception Distance, a standard metric for generative models.\n  We show that this apparent gap vanishes when denoising the final Langevin"},"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":"2009.05475","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-09-11T14:49:53Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"66e8d661d26386f198880c87c513360834c2d66ad62ab6d725b564a995078777","abstract_canon_sha256":"fb3929feeca5a46af08708e6836939de67db9972e2a98cc27e11c080e6280568"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:41:57.130038Z","signature_b64":"pkRQTOdHRY7uAy74QxISDRg7K9OOhXx5AcXBGQwEGK5WstH3BrKJNXZZgEB4LHtdz0an+HLV+DPMb44z3YUmDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3514208115c23dc996ecfc58fa8548d120163ddd57dc7f77b865bce166dd1a4d","last_reissued_at":"2026-07-05T01:41:57.129593Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:41:57.129593Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Adversarial score matching and improved sampling for image generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexia Jolicoeur-Martineau, Ioannis Mitliagkas, R\\'emi Pich\\'e-Taillefer, R\\'emi Tachet des Combes","submitted_at":"2020-09-11T14:49:53Z","abstract_excerpt":"Denoising Score Matching with Annealed Langevin Sampling (DSM-ALS) has recently found success in generative modeling. The approach works by first training a neural network to estimate the score of a distribution, and then using Langevin dynamics to sample from the data distribution assumed by the score network. Despite the convincing visual quality of samples, this method appears to perform worse than Generative Adversarial Networks (GANs) under the Fr\\'echet Inception Distance, a standard metric for generative models.\n  We show that this apparent gap vanishes when denoising the final Langevin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.05475","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2009.05475/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":"2009.05475","created_at":"2026-07-05T01:41:57.129650+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.05475v2","created_at":"2026-07-05T01:41:57.129650+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.05475","created_at":"2026-07-05T01:41:57.129650+00:00"},{"alias_kind":"pith_short_12","alias_value":"GUKCBAIVYI64","created_at":"2026-07-05T01:41:57.129650+00:00"},{"alias_kind":"pith_short_16","alias_value":"GUKCBAIVYI64TFXM","created_at":"2026-07-05T01:41:57.129650+00:00"},{"alias_kind":"pith_short_8","alias_value":"GUKCBAIV","created_at":"2026-07-05T01:41:57.129650+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19162","citing_title":"The Reward Was in Your Data All Along: Correcting Flow Matching with Discriminator-Guided RL","ref_index":116,"is_internal_anchor":false},{"citing_arxiv_id":"2011.13456","citing_title":"Score-Based Generative Modeling through Stochastic Differential Equations","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2105.05233","citing_title":"Diffusion Models Beat GANs on Image Synthesis","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00229","citing_title":"A unified perspective on fine-tuning and sampling with diffusion and flow models","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E","json":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E.json","graph_json":"https://pith.science/api/pith-number/GUKCBAIVYI64TFXM7RMPVBKI2E/graph.json","events_json":"https://pith.science/api/pith-number/GUKCBAIVYI64TFXM7RMPVBKI2E/events.json","paper":"https://pith.science/paper/GUKCBAIV"},"agent_actions":{"view_html":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E","download_json":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E.json","view_paper":"https://pith.science/paper/GUKCBAIV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.05475&json=true","fetch_graph":"https://pith.science/api/pith-number/GUKCBAIVYI64TFXM7RMPVBKI2E/graph.json","fetch_events":"https://pith.science/api/pith-number/GUKCBAIVYI64TFXM7RMPVBKI2E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E/action/storage_attestation","attest_author":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E/action/author_attestation","sign_citation":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E/action/citation_signature","submit_replication":"https://pith.science/pith/GUKCBAIVYI64TFXM7RMPVBKI2E/action/replication_record"}},"created_at":"2026-07-05T01:41:57.129650+00:00","updated_at":"2026-07-05T01:41:57.129650+00:00"}