{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WYC3NXYRIYOWLEWFBCOBRRZ4OK","short_pith_number":"pith:WYC3NXYR","schema_version":"1.0","canonical_sha256":"b605b6df11461d6592c5089c18c73c728aaf41bea2a8eb808d3abccc471a51fb","source":{"kind":"arxiv","id":"2507.04619","version":1},"attestation_state":"computed","paper":{"title":"Information-Guided Diffusion Sampling for Dataset Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Guang Li, Konstantinos N. Plataniotis, Linfeng Ye, Miki Haseyama, Shayan Mohajer Hamidi, Takahiro Ogawa","submitted_at":"2025-07-07T02:27:08Z","abstract_excerpt":"Dataset distillation aims to create a compact dataset that retains essential information while maintaining model performance. Diffusion models (DMs) have shown promise for this task but struggle in low images-per-class (IPC) settings, where generated samples lack diversity. In this paper, we address this issue from an information-theoretic perspective by identifying two key types of information that a distilled dataset must preserve: ($i$) prototype information $\\mathrm{I}(X;Y)$, which captures label-relevant features; and ($ii$) contextual information $\\mathrm{H}(X | Y)$, which preserves intr"},"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":"2507.04619","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-07T02:27:08Z","cross_cats_sorted":["cs.AI","cs.CV","cs.IT","math.IT"],"title_canon_sha256":"d35c42a167ee3d7bd8d5a99bdd4bfbf41038b9decb09df39202f8225f231d1b0","abstract_canon_sha256":"36909387432ca3612c0d3088feb80d370502f8896240ed73445413973ad1920c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:50.769472Z","signature_b64":"3OlpsRSpuQJtd1o01jEiezwbPd9TuU95EuQtlea2x+As/CEnJLddF1Ga9n6zd3ZVisE0/t8X/8bmmdRdXUO5CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b605b6df11461d6592c5089c18c73c728aaf41bea2a8eb808d3abccc471a51fb","last_reissued_at":"2026-07-05T11:32:50.768974Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:50.768974Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Information-Guided Diffusion Sampling for Dataset Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.IT","math.IT"],"primary_cat":"cs.LG","authors_text":"Guang Li, Konstantinos N. Plataniotis, Linfeng Ye, Miki Haseyama, Shayan Mohajer Hamidi, Takahiro Ogawa","submitted_at":"2025-07-07T02:27:08Z","abstract_excerpt":"Dataset distillation aims to create a compact dataset that retains essential information while maintaining model performance. Diffusion models (DMs) have shown promise for this task but struggle in low images-per-class (IPC) settings, where generated samples lack diversity. In this paper, we address this issue from an information-theoretic perspective by identifying two key types of information that a distilled dataset must preserve: ($i$) prototype information $\\mathrm{I}(X;Y)$, which captures label-relevant features; and ($ii$) contextual information $\\mathrm{H}(X | Y)$, which preserves intr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.04619","kind":"arxiv","version":1},"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/2507.04619/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":"2507.04619","created_at":"2026-07-05T11:32:50.769032+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.04619v1","created_at":"2026-07-05T11:32:50.769032+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.04619","created_at":"2026-07-05T11:32:50.769032+00:00"},{"alias_kind":"pith_short_12","alias_value":"WYC3NXYRIYOW","created_at":"2026-07-05T11:32:50.769032+00:00"},{"alias_kind":"pith_short_16","alias_value":"WYC3NXYRIYOWLEWF","created_at":"2026-07-05T11:32:50.769032+00:00"},{"alias_kind":"pith_short_8","alias_value":"WYC3NXYR","created_at":"2026-07-05T11:32:50.769032+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01920","citing_title":"Pool-Select-Refine for Allocation-Aware Generative Dataset Distillation","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK","json":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK.json","graph_json":"https://pith.science/api/pith-number/WYC3NXYRIYOWLEWFBCOBRRZ4OK/graph.json","events_json":"https://pith.science/api/pith-number/WYC3NXYRIYOWLEWFBCOBRRZ4OK/events.json","paper":"https://pith.science/paper/WYC3NXYR"},"agent_actions":{"view_html":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK","download_json":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK.json","view_paper":"https://pith.science/paper/WYC3NXYR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.04619&json=true","fetch_graph":"https://pith.science/api/pith-number/WYC3NXYRIYOWLEWFBCOBRRZ4OK/graph.json","fetch_events":"https://pith.science/api/pith-number/WYC3NXYRIYOWLEWFBCOBRRZ4OK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK/action/storage_attestation","attest_author":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK/action/author_attestation","sign_citation":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK/action/citation_signature","submit_replication":"https://pith.science/pith/WYC3NXYRIYOWLEWFBCOBRRZ4OK/action/replication_record"}},"created_at":"2026-07-05T11:32:50.769032+00:00","updated_at":"2026-07-05T11:32:50.769032+00:00"}