{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:Q6BVK353NFV5UPT55S6E2FGLCQ","short_pith_number":"pith:Q6BVK353","schema_version":"1.0","canonical_sha256":"8783556fbb696bda3e7decbc4d14cb1415246cb691080d5260a534ecc308acd3","source":{"kind":"arxiv","id":"2310.17347","version":4},"attestation_state":"computed","paper":{"title":"CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Sampling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Derek Bradley, Jakob Buhmann, Otmar Hilliges, Romann M. Weber, Seyedmorteza Sadat","submitted_at":"2023-10-26T12:27:56Z","abstract_excerpt":"While conditional diffusion models are known to have good coverage of the data distribution, they still face limitations in output diversity, particularly when sampled with a high classifier-free guidance scale for optimal image quality or when trained on small datasets. We attribute this problem to the role of the conditioning signal in inference and offer an improved sampling strategy for diffusion models that can increase generation diversity, especially at high guidance scales, with minimal loss of sample quality. Our sampling strategy anneals the conditioning signal by adding scheduled, m"},"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":"2310.17347","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-10-26T12:27:56Z","cross_cats_sorted":[],"title_canon_sha256":"71940b86d639438c4312731bcbe6b14aba14edc328808db31f1c7cf97b287d24","abstract_canon_sha256":"4b217de020141464ce589afabbeae81cb43b53edd49d83205e60a7e7cf96d36b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:18:32.390866Z","signature_b64":"FnrQOGY4vOByaiRELclKDDZU5jDf4OV8tCCWseTENMdnFa3PAwl5yO8EEwv+9gOTkII8/8PNKEFu3Tqc9teeBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8783556fbb696bda3e7decbc4d14cb1415246cb691080d5260a534ecc308acd3","last_reissued_at":"2026-07-05T08:18:32.390388Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:18:32.390388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Sampling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Derek Bradley, Jakob Buhmann, Otmar Hilliges, Romann M. Weber, Seyedmorteza Sadat","submitted_at":"2023-10-26T12:27:56Z","abstract_excerpt":"While conditional diffusion models are known to have good coverage of the data distribution, they still face limitations in output diversity, particularly when sampled with a high classifier-free guidance scale for optimal image quality or when trained on small datasets. We attribute this problem to the role of the conditioning signal in inference and offer an improved sampling strategy for diffusion models that can increase generation diversity, especially at high guidance scales, with minimal loss of sample quality. Our sampling strategy anneals the conditioning signal by adding scheduled, m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.17347","kind":"arxiv","version":4},"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/2310.17347/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":"2310.17347","created_at":"2026-07-05T08:18:32.390443+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.17347v4","created_at":"2026-07-05T08:18:32.390443+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.17347","created_at":"2026-07-05T08:18:32.390443+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q6BVK353NFV5","created_at":"2026-07-05T08:18:32.390443+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q6BVK353NFV5UPT5","created_at":"2026-07-05T08:18:32.390443+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q6BVK353","created_at":"2026-07-05T08:18:32.390443+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27371","citing_title":"Don't Settle at the Mode! Mitigating Diversity Collapse in Pretrained Flow Models via Feature Self-Guidance","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23679","citing_title":"Semantic Browsing: Controllable Diversity for Image Generation","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06813","citing_title":"Breaking the Lock-in: Diversifying Text-to-Image Generation via Representation Modulation","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2507.02314","citing_title":"MAGIC: Few-Shot Mask-Guided Anomaly Inpainting with Prompt Perturbation, Spatially Adaptive Guidance, and Context Awareness","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2601.00090","citing_title":"It's Never Too Late: Noise Optimization for Collapse Recovery in Trained Diffusion Models","ref_index":53,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11494","citing_title":"STRIDE: Training-Free Diversity Guidance via PCA-Directed Feature Perturbation in Single-Step Diffusion Models","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03877","citing_title":"DMGD: Train-Free Dataset Distillation with Semantic-Distribution Matching in Diffusion Models","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17323","citing_title":"A Universal Avoidance Method for Diverse Multi-branch Generation","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ","json":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ.json","graph_json":"https://pith.science/api/pith-number/Q6BVK353NFV5UPT55S6E2FGLCQ/graph.json","events_json":"https://pith.science/api/pith-number/Q6BVK353NFV5UPT55S6E2FGLCQ/events.json","paper":"https://pith.science/paper/Q6BVK353"},"agent_actions":{"view_html":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ","download_json":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ.json","view_paper":"https://pith.science/paper/Q6BVK353","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.17347&json=true","fetch_graph":"https://pith.science/api/pith-number/Q6BVK353NFV5UPT55S6E2FGLCQ/graph.json","fetch_events":"https://pith.science/api/pith-number/Q6BVK353NFV5UPT55S6E2FGLCQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ/action/storage_attestation","attest_author":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ/action/author_attestation","sign_citation":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ/action/citation_signature","submit_replication":"https://pith.science/pith/Q6BVK353NFV5UPT55S6E2FGLCQ/action/replication_record"}},"created_at":"2026-07-05T08:18:32.390443+00:00","updated_at":"2026-07-05T08:18:32.390443+00:00"}