{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:EFA5WFHBWS5F7PDJ7R5XB45BZ7","short_pith_number":"pith:EFA5WFHB","schema_version":"1.0","canonical_sha256":"2141db14e1b4ba5fbc69fc7b70f3a1cff4ff4990ccd9a0639dc6bb3bd77a5f37","source":{"kind":"arxiv","id":"2203.17260","version":2},"attestation_state":"computed","paper":{"title":"Generating High Fidelity Data from Low-density Regions using Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Albert Gordo, Caner Hazirbas, Cristian Canton Ferrer, Firat Ozgenel, Vikash Sehwag","submitted_at":"2022-03-31T17:56:25Z","abstract_excerpt":"Our work focuses on addressing sample deficiency from low-density regions of data manifold in common image datasets. We leverage diffusion process based generative models to synthesize novel images from low-density regions. We observe that uniform sampling from diffusion models predominantly samples from high-density regions of the data manifold. Therefore, we modify the sampling process to guide it towards low-density regions while simultaneously maintaining the fidelity of synthetic data. We rigorously demonstrate that our process successfully generates novel high fidelity samples from low-d"},"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":"2203.17260","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-03-31T17:56:25Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"41f56dc229581c1fcdb14492ef5bfde82140762bd0cb23ff02d77bca057ffb6f","abstract_canon_sha256":"00027c2ff23eea47f80f4558dac7fff945793c76f09f638d9fdee3875b2eb470"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:34:52.569434Z","signature_b64":"11SC7Y9lt5rrfmognfq4M0bjvX3xThqkpc5Va5SK8j+ajdAtTcmgmgGYUj3cMPBHmJIv5XlrIv3poZxZztVDDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2141db14e1b4ba5fbc69fc7b70f3a1cff4ff4990ccd9a0639dc6bb3bd77a5f37","last_reissued_at":"2026-07-05T04:34:52.568933Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:34:52.568933Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generating High Fidelity Data from Low-density Regions using Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Albert Gordo, Caner Hazirbas, Cristian Canton Ferrer, Firat Ozgenel, Vikash Sehwag","submitted_at":"2022-03-31T17:56:25Z","abstract_excerpt":"Our work focuses on addressing sample deficiency from low-density regions of data manifold in common image datasets. We leverage diffusion process based generative models to synthesize novel images from low-density regions. We observe that uniform sampling from diffusion models predominantly samples from high-density regions of the data manifold. Therefore, we modify the sampling process to guide it towards low-density regions while simultaneously maintaining the fidelity of synthetic data. We rigorously demonstrate that our process successfully generates novel high fidelity samples from low-d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.17260","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/2203.17260/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":"2203.17260","created_at":"2026-07-05T04:34:52.568992+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.17260v2","created_at":"2026-07-05T04:34:52.568992+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.17260","created_at":"2026-07-05T04:34:52.568992+00:00"},{"alias_kind":"pith_short_12","alias_value":"EFA5WFHBWS5F","created_at":"2026-07-05T04:34:52.568992+00:00"},{"alias_kind":"pith_short_16","alias_value":"EFA5WFHBWS5F7PDJ","created_at":"2026-07-05T04:34:52.568992+00:00"},{"alias_kind":"pith_short_8","alias_value":"EFA5WFHB","created_at":"2026-07-05T04:34:52.568992+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7","json":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7.json","graph_json":"https://pith.science/api/pith-number/EFA5WFHBWS5F7PDJ7R5XB45BZ7/graph.json","events_json":"https://pith.science/api/pith-number/EFA5WFHBWS5F7PDJ7R5XB45BZ7/events.json","paper":"https://pith.science/paper/EFA5WFHB"},"agent_actions":{"view_html":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7","download_json":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7.json","view_paper":"https://pith.science/paper/EFA5WFHB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.17260&json=true","fetch_graph":"https://pith.science/api/pith-number/EFA5WFHBWS5F7PDJ7R5XB45BZ7/graph.json","fetch_events":"https://pith.science/api/pith-number/EFA5WFHBWS5F7PDJ7R5XB45BZ7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7/action/storage_attestation","attest_author":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7/action/author_attestation","sign_citation":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7/action/citation_signature","submit_replication":"https://pith.science/pith/EFA5WFHBWS5F7PDJ7R5XB45BZ7/action/replication_record"}},"created_at":"2026-07-05T04:34:52.568992+00:00","updated_at":"2026-07-05T04:34:52.568992+00:00"}