{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:K4RGLFRIMTOWACACSLT55HF2LO","short_pith_number":"pith:K4RGLFRI","schema_version":"1.0","canonical_sha256":"572265962864dd60080292e7de9cba5ba47db39c659c749ae0f3c374de217c4c","source":{"kind":"arxiv","id":"2502.13747","version":2},"attestation_state":"computed","paper":{"title":"Reverse Markov Learning: Multi-Step Generative Models for Complex Distributions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ME","stat.ML"],"primary_cat":"cs.LG","authors_text":"Nicolai Meinshausen, Tong Zhang, Xinwei Shen","submitted_at":"2025-02-19T14:10:15Z","abstract_excerpt":"Learning complex distributions is a fundamental challenge in contemporary applications. Shen and Meinshausen (2024) introduced engression, a generative approach based on scoring rules that maps noise (and covariates, if available) directly to data. While effective, engression can struggle with highly complex distributions, such as those encountered in image data. In this work, we propose reverse Markov learning (RML), a framework that defines a general forward process transitioning from the target distribution to a known distribution (e.g., Gaussian) and then learns a reverse Markov process us"},"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":"2502.13747","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T14:10:15Z","cross_cats_sorted":["stat.ME","stat.ML"],"title_canon_sha256":"9c8b906da8d8fc64d39aa278196537b828ca47ad7ef5b5023fc442539c0332a3","abstract_canon_sha256":"4a32bfae06edc3d35c48509e6bf4218ae75eeffd00b14e85ae888711ef2c6d1f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:55:16.851525Z","signature_b64":"MnyEh2JMJbhHLZeIzwS9UYQfyJOZWlpVN6hiy4T0O8tXILDP797+HJ+U+kn/BwskhBgRoBekDquA9ldd0aHBAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"572265962864dd60080292e7de9cba5ba47db39c659c749ae0f3c374de217c4c","last_reissued_at":"2026-07-05T11:55:16.851104Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:55:16.851104Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reverse Markov Learning: Multi-Step Generative Models for Complex Distributions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ME","stat.ML"],"primary_cat":"cs.LG","authors_text":"Nicolai Meinshausen, Tong Zhang, Xinwei Shen","submitted_at":"2025-02-19T14:10:15Z","abstract_excerpt":"Learning complex distributions is a fundamental challenge in contemporary applications. Shen and Meinshausen (2024) introduced engression, a generative approach based on scoring rules that maps noise (and covariates, if available) directly to data. While effective, engression can struggle with highly complex distributions, such as those encountered in image data. In this work, we propose reverse Markov learning (RML), a framework that defines a general forward process transitioning from the target distribution to a known distribution (e.g., Gaussian) and then learns a reverse Markov process us"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.13747","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/2502.13747/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":"2502.13747","created_at":"2026-07-05T11:55:16.851159+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.13747v2","created_at":"2026-07-05T11:55:16.851159+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.13747","created_at":"2026-07-05T11:55:16.851159+00:00"},{"alias_kind":"pith_short_12","alias_value":"K4RGLFRIMTOW","created_at":"2026-07-05T11:55:16.851159+00:00"},{"alias_kind":"pith_short_16","alias_value":"K4RGLFRIMTOWACAC","created_at":"2026-07-05T11:55:16.851159+00:00"},{"alias_kind":"pith_short_8","alias_value":"K4RGLFRI","created_at":"2026-07-05T11:55:16.851159+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2502.11583","citing_title":"Distributional Autoencoders Know the Score","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2509.26258","citing_title":"EnScale: Temporally-consistent multivariate generative downscaling via proper scoring rules","ref_index":51,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO","json":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO.json","graph_json":"https://pith.science/api/pith-number/K4RGLFRIMTOWACACSLT55HF2LO/graph.json","events_json":"https://pith.science/api/pith-number/K4RGLFRIMTOWACACSLT55HF2LO/events.json","paper":"https://pith.science/paper/K4RGLFRI"},"agent_actions":{"view_html":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO","download_json":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO.json","view_paper":"https://pith.science/paper/K4RGLFRI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.13747&json=true","fetch_graph":"https://pith.science/api/pith-number/K4RGLFRIMTOWACACSLT55HF2LO/graph.json","fetch_events":"https://pith.science/api/pith-number/K4RGLFRIMTOWACACSLT55HF2LO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO/action/storage_attestation","attest_author":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO/action/author_attestation","sign_citation":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO/action/citation_signature","submit_replication":"https://pith.science/pith/K4RGLFRIMTOWACACSLT55HF2LO/action/replication_record"}},"created_at":"2026-07-05T11:55:16.851159+00:00","updated_at":"2026-07-05T11:55:16.851159+00:00"}