{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:JLRIXX7UNL4EQHHE4US5AMBRNQ","short_pith_number":"pith:JLRIXX7U","schema_version":"1.0","canonical_sha256":"4ae28bdff46af8481ce4e525d030316c112c508ba6f5d0557d04d98a5fe35ee1","source":{"kind":"arxiv","id":"2311.07604","version":2},"attestation_state":"computed","paper":{"title":"Finetuning Text-to-Image Diffusion Models for Fairness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.CY"],"primary_cat":"cs.LG","authors_text":"Chao Du, Min Lin, Mohan Kankanhalli, Tianyu Pang, Xudong Shen, Yongkang Wong","submitted_at":"2023-11-11T05:40:54Z","abstract_excerpt":"The rapid adoption of text-to-image diffusion models in society underscores an urgent need to address their biases. Without interventions, these biases could propagate a skewed worldview and restrict opportunities for minority groups. In this work, we frame fairness as a distributional alignment problem. Our solution consists of two main technical contributions: (1) a distributional alignment loss that steers specific characteristics of the generated images towards a user-defined target distribution, and (2) adjusted direct finetuning of diffusion model's sampling process (adjusted DFT), which"},"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":"2311.07604","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-11-11T05:40:54Z","cross_cats_sorted":["cs.AI","cs.CV","cs.CY"],"title_canon_sha256":"0c7a5396f46ccf11a042549128d3424d00dd747654c872a75743d807a414642a","abstract_canon_sha256":"d6e9ff8f490288c5e164c8d1536dce0491c94f1352e763db89c3cffc841fc08e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:56:18.849216Z","signature_b64":"Yigc5G/31FJqsR34wslCbS0FZMUiMcT0Gl+zWqlUNQkIbWKlQqVin2BAMPpaTp5+TawEDOenOsoqnDC5ncs6Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ae28bdff46af8481ce4e525d030316c112c508ba6f5d0557d04d98a5fe35ee1","last_reissued_at":"2026-07-05T07:56:18.848737Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:56:18.848737Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Finetuning Text-to-Image Diffusion Models for Fairness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.CY"],"primary_cat":"cs.LG","authors_text":"Chao Du, Min Lin, Mohan Kankanhalli, Tianyu Pang, Xudong Shen, Yongkang Wong","submitted_at":"2023-11-11T05:40:54Z","abstract_excerpt":"The rapid adoption of text-to-image diffusion models in society underscores an urgent need to address their biases. Without interventions, these biases could propagate a skewed worldview and restrict opportunities for minority groups. In this work, we frame fairness as a distributional alignment problem. Our solution consists of two main technical contributions: (1) a distributional alignment loss that steers specific characteristics of the generated images towards a user-defined target distribution, and (2) adjusted direct finetuning of diffusion model's sampling process (adjusted DFT), which"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.07604","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/2311.07604/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":"2311.07604","created_at":"2026-07-05T07:56:18.848804+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.07604v2","created_at":"2026-07-05T07:56:18.848804+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.07604","created_at":"2026-07-05T07:56:18.848804+00:00"},{"alias_kind":"pith_short_12","alias_value":"JLRIXX7UNL4E","created_at":"2026-07-05T07:56:18.848804+00:00"},{"alias_kind":"pith_short_16","alias_value":"JLRIXX7UNL4EQHHE","created_at":"2026-07-05T07:56:18.848804+00:00"},{"alias_kind":"pith_short_8","alias_value":"JLRIXX7U","created_at":"2026-07-05T07:56:18.848804+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"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":151,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01147","citing_title":"EquiSteer: Cross-Attention Steering Towards a Fairer Text-Guided Image Generation","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00423","citing_title":"Selective Test-Time Debiasing for CLIP via Reward Gating","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00377","citing_title":"Score-Control for Hallucination Reduction in Diffusion Models","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30262","citing_title":"Intermediate Text Representation Guided Text-to-Image Generation for Enhancing One-and-Only Alignment","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06052","citing_title":"Attention, May I Have Your Decision? Localizing Generative Choices in Diffusion Models","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18167","citing_title":"Embedding Arithmetic: A Lightweight, Tuning-Free Framework for Post-hoc Bias Mitigation in Text-to-Image Models","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ","json":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ.json","graph_json":"https://pith.science/api/pith-number/JLRIXX7UNL4EQHHE4US5AMBRNQ/graph.json","events_json":"https://pith.science/api/pith-number/JLRIXX7UNL4EQHHE4US5AMBRNQ/events.json","paper":"https://pith.science/paper/JLRIXX7U"},"agent_actions":{"view_html":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ","download_json":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ.json","view_paper":"https://pith.science/paper/JLRIXX7U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.07604&json=true","fetch_graph":"https://pith.science/api/pith-number/JLRIXX7UNL4EQHHE4US5AMBRNQ/graph.json","fetch_events":"https://pith.science/api/pith-number/JLRIXX7UNL4EQHHE4US5AMBRNQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ/action/storage_attestation","attest_author":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ/action/author_attestation","sign_citation":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ/action/citation_signature","submit_replication":"https://pith.science/pith/JLRIXX7UNL4EQHHE4US5AMBRNQ/action/replication_record"}},"created_at":"2026-07-05T07:56:18.848804+00:00","updated_at":"2026-07-05T07:56:18.848804+00:00"}