{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:SE2U7FLPCG7K7NBSWE6XW6PUO6","short_pith_number":"pith:SE2U7FLP","schema_version":"1.0","canonical_sha256":"91354f956f11beafb432b13d7b79f477a94bfb8631b7b2fa604945a2bf7aee9c","source":{"kind":"arxiv","id":"2304.11961","version":3},"attestation_state":"computed","paper":{"title":"Towards Mode Balancing of Generative Models via Diversity Weights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Christian Guckelsberger, Sebastian Berns, Simon Colton","submitted_at":"2023-04-24T09:55:17Z","abstract_excerpt":"Large data-driven image models are extensively used to support creative and artistic work. Under the currently predominant distribution-fitting paradigm, a dataset is treated as ground truth to be approximated as closely as possible. Yet, many creative applications demand a diverse range of output, and creators often strive to actively diverge from a given data distribution. We argue that an adjustment of modelling objectives, from pure mode coverage towards mode balancing, is necessary to accommodate the goal of higher output diversity. We present diversity weights, a training scheme that inc"},"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":"2304.11961","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-04-24T09:55:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9343e20fe85e5df49e51cbf51f3c4f007d459edc705508ae781461f5eb3c5ece","abstract_canon_sha256":"a33e643aac10802594e66f34facd576242a12754b56fa3fb9eccbe4f15e5ef36"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:21:06.131716Z","signature_b64":"gV6EOk1GSI/mjYzr72mGHdtTKYsAk8xkyMbXUNsoin+zI84NOxOvHPDAb26VKmLtpYHT8ytwdysI9jvsAcEIBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"91354f956f11beafb432b13d7b79f477a94bfb8631b7b2fa604945a2bf7aee9c","last_reissued_at":"2026-07-05T06:21:06.131241Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:21:06.131241Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Mode Balancing of Generative Models via Diversity Weights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Christian Guckelsberger, Sebastian Berns, Simon Colton","submitted_at":"2023-04-24T09:55:17Z","abstract_excerpt":"Large data-driven image models are extensively used to support creative and artistic work. Under the currently predominant distribution-fitting paradigm, a dataset is treated as ground truth to be approximated as closely as possible. Yet, many creative applications demand a diverse range of output, and creators often strive to actively diverge from a given data distribution. We argue that an adjustment of modelling objectives, from pure mode coverage towards mode balancing, is necessary to accommodate the goal of higher output diversity. We present diversity weights, a training scheme that inc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.11961","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/2304.11961/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":"2304.11961","created_at":"2026-07-05T06:21:06.131298+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.11961v3","created_at":"2026-07-05T06:21:06.131298+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.11961","created_at":"2026-07-05T06:21:06.131298+00:00"},{"alias_kind":"pith_short_12","alias_value":"SE2U7FLPCG7K","created_at":"2026-07-05T06:21:06.131298+00:00"},{"alias_kind":"pith_short_16","alias_value":"SE2U7FLPCG7K7NBS","created_at":"2026-07-05T06:21:06.131298+00:00"},{"alias_kind":"pith_short_8","alias_value":"SE2U7FLP","created_at":"2026-07-05T06:21:06.131298+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.04593","citing_title":"The Alpha-Alternator: Dynamic Adaptation To Varying Noise Levels In Sequences Using The Vendi Score For Improved Robustness and Performance","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6","json":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6.json","graph_json":"https://pith.science/api/pith-number/SE2U7FLPCG7K7NBSWE6XW6PUO6/graph.json","events_json":"https://pith.science/api/pith-number/SE2U7FLPCG7K7NBSWE6XW6PUO6/events.json","paper":"https://pith.science/paper/SE2U7FLP"},"agent_actions":{"view_html":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6","download_json":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6.json","view_paper":"https://pith.science/paper/SE2U7FLP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.11961&json=true","fetch_graph":"https://pith.science/api/pith-number/SE2U7FLPCG7K7NBSWE6XW6PUO6/graph.json","fetch_events":"https://pith.science/api/pith-number/SE2U7FLPCG7K7NBSWE6XW6PUO6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6/action/storage_attestation","attest_author":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6/action/author_attestation","sign_citation":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6/action/citation_signature","submit_replication":"https://pith.science/pith/SE2U7FLPCG7K7NBSWE6XW6PUO6/action/replication_record"}},"created_at":"2026-07-05T06:21:06.131298+00:00","updated_at":"2026-07-05T06:21:06.131298+00:00"}