{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:DISKRRNADQDO4TO4Z2BRZ43ICQ","short_pith_number":"pith:DISKRRNA","schema_version":"1.0","canonical_sha256":"1a24a8c5a01c06ee4ddcce831cf3681419d87ce33fb5cd2abd182ba725e5c3a1","source":{"kind":"arxiv","id":"2302.14082","version":2},"attestation_state":"computed","paper":{"title":"Detecting and Mitigating Mode-Collapse for Flow-based Sampling of Lattice Field Theories","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","physics.comp-ph"],"primary_cat":"hep-lat","authors_text":"Christopher J. Anders, Karl Jansen, Kim A. Nicoli, Pan Kessel, Shinichi Nakajima, Tobias Hartung","submitted_at":"2023-02-27T19:00:22Z","abstract_excerpt":"We study the consequences of mode-collapse of normalizing flows in the context of lattice field theory. Normalizing flows allow for independent sampling. For this reason, it is hoped that they can avoid the tunneling problem of local-update MCMC algorithms for multi-modal distributions. In this work, we first point out that the tunneling problem is also present for normalizing flows but is shifted from the sampling to the training phase of the algorithm. Specifically, normalizing flows often suffer from mode-collapse for which the training process assigns vanishingly low probability mass to re"},"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":"2302.14082","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"hep-lat","submitted_at":"2023-02-27T19:00:22Z","cross_cats_sorted":["cs.LG","physics.comp-ph"],"title_canon_sha256":"9a91bf618fa88fd033d77bae5ffe20add285f002840e5790dd22fb917514766a","abstract_canon_sha256":"2d22e364fa0d153179d9e440f67203aae1b10246de193c16f2e42e9e5fa8f31f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:08:40.884700Z","signature_b64":"Ng/9K65RWiCOeNtBV/8E79vxqDr8+WY0+UoDcwGb7pLhcqWLYsb3YOToxtNzn/ljpW80Z1LfRQnP4xCr5rCZDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a24a8c5a01c06ee4ddcce831cf3681419d87ce33fb5cd2abd182ba725e5c3a1","last_reissued_at":"2026-07-05T07:08:40.884205Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:08:40.884205Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Detecting and Mitigating Mode-Collapse for Flow-based Sampling of Lattice Field Theories","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","physics.comp-ph"],"primary_cat":"hep-lat","authors_text":"Christopher J. Anders, Karl Jansen, Kim A. Nicoli, Pan Kessel, Shinichi Nakajima, Tobias Hartung","submitted_at":"2023-02-27T19:00:22Z","abstract_excerpt":"We study the consequences of mode-collapse of normalizing flows in the context of lattice field theory. Normalizing flows allow for independent sampling. For this reason, it is hoped that they can avoid the tunneling problem of local-update MCMC algorithms for multi-modal distributions. In this work, we first point out that the tunneling problem is also present for normalizing flows but is shifted from the sampling to the training phase of the algorithm. Specifically, normalizing flows often suffer from mode-collapse for which the training process assigns vanishingly low probability mass to re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.14082","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/2302.14082/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":"2302.14082","created_at":"2026-07-05T07:08:40.884262+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.14082v2","created_at":"2026-07-05T07:08:40.884262+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.14082","created_at":"2026-07-05T07:08:40.884262+00:00"},{"alias_kind":"pith_short_12","alias_value":"DISKRRNADQDO","created_at":"2026-07-05T07:08:40.884262+00:00"},{"alias_kind":"pith_short_16","alias_value":"DISKRRNADQDO4TO4","created_at":"2026-07-05T07:08:40.884262+00:00"},{"alias_kind":"pith_short_8","alias_value":"DISKRRNA","created_at":"2026-07-05T07:08:40.884262+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.07262","citing_title":"Testing machine-learned distributions against Monte Carlo data for the QCD chiral phase transition","ref_index":55,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ","json":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ.json","graph_json":"https://pith.science/api/pith-number/DISKRRNADQDO4TO4Z2BRZ43ICQ/graph.json","events_json":"https://pith.science/api/pith-number/DISKRRNADQDO4TO4Z2BRZ43ICQ/events.json","paper":"https://pith.science/paper/DISKRRNA"},"agent_actions":{"view_html":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ","download_json":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ.json","view_paper":"https://pith.science/paper/DISKRRNA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.14082&json=true","fetch_graph":"https://pith.science/api/pith-number/DISKRRNADQDO4TO4Z2BRZ43ICQ/graph.json","fetch_events":"https://pith.science/api/pith-number/DISKRRNADQDO4TO4Z2BRZ43ICQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ/action/storage_attestation","attest_author":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ/action/author_attestation","sign_citation":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ/action/citation_signature","submit_replication":"https://pith.science/pith/DISKRRNADQDO4TO4Z2BRZ43ICQ/action/replication_record"}},"created_at":"2026-07-05T07:08:40.884262+00:00","updated_at":"2026-07-05T07:08:40.884262+00:00"}