{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:CMBUID3PCY7KJDJDM6N4OLXWUZ","short_pith_number":"pith:CMBUID3P","schema_version":"1.0","canonical_sha256":"1303440f6f163ea48d23679bc72ef6a664195e7b22416f103aa1612913580609","source":{"kind":"arxiv","id":"2006.05929","version":3},"attestation_state":"computed","paper":{"title":"Dataset Condensation with Gradient Matching","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Zhao, Hakan Bilen, Konda Reddy Mopuri","submitted_at":"2020-06-10T16:30:52Z","abstract_excerpt":"As the state-of-the-art machine learning methods in many fields rely on larger datasets, storing datasets and training models on them become significantly more expensive. This paper proposes a training set synthesis technique for data-efficient learning, called Dataset Condensation, that learns to condense large dataset into a small set of informative synthetic samples for training deep neural networks from scratch. We formulate this goal as a gradient matching problem between the gradients of deep neural network weights that are trained on the original and our synthetic data. We rigorously ev"},"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":"2006.05929","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2020-06-10T16:30:52Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6f5b7c3fc21dcc6a497f141a0593c3c89b2457b0f45e51b3aaacb1a644461102","abstract_canon_sha256":"4d144bff3949a61cfe12193447dc211b243ffc1966ea8f2da62c1e263de90195"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:20:55.211048Z","signature_b64":"Wyx5s+JDU7eDQQm+Cp1Xs1RGVakSXndBgEku+M5JIr787Bvuzx8q64qIuvErhGtJzSL2/dBa5UcgFmjqOy32DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1303440f6f163ea48d23679bc72ef6a664195e7b22416f103aa1612913580609","last_reissued_at":"2026-07-05T02:20:55.210584Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:20:55.210584Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dataset Condensation with Gradient Matching","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Bo Zhao, Hakan Bilen, Konda Reddy Mopuri","submitted_at":"2020-06-10T16:30:52Z","abstract_excerpt":"As the state-of-the-art machine learning methods in many fields rely on larger datasets, storing datasets and training models on them become significantly more expensive. This paper proposes a training set synthesis technique for data-efficient learning, called Dataset Condensation, that learns to condense large dataset into a small set of informative synthetic samples for training deep neural networks from scratch. We formulate this goal as a gradient matching problem between the gradients of deep neural network weights that are trained on the original and our synthetic data. We rigorously ev"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.05929","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/2006.05929/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":"2006.05929","created_at":"2026-07-05T02:20:55.210644+00:00"},{"alias_kind":"arxiv_version","alias_value":"2006.05929v3","created_at":"2026-07-05T02:20:55.210644+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.05929","created_at":"2026-07-05T02:20:55.210644+00:00"},{"alias_kind":"pith_short_12","alias_value":"CMBUID3PCY7K","created_at":"2026-07-05T02:20:55.210644+00:00"},{"alias_kind":"pith_short_16","alias_value":"CMBUID3PCY7KJDJD","created_at":"2026-07-05T02:20:55.210644+00:00"},{"alias_kind":"pith_short_8","alias_value":"CMBUID3P","created_at":"2026-07-05T02:20:55.210644+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":16,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.20196","citing_title":"Distill Once, Adapt Life-Long: Exploring Dataset Distillation for Continual Test-Time Adaptation","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00916","citing_title":"Condensing Large-Scale Datasets Directly with Minimal Information Loss","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01920","citing_title":"Pool-Select-Refine for Allocation-Aware Generative Dataset Distillation","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20196","citing_title":"Distill Once, Adapt Life-Long: Exploring Dataset Distillation for Continual Test-Time Adaptation","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29464","citing_title":"Rank-Aware Hyperbolic Alignment for Vision-Language Dataset Distillation","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29837","citing_title":"Robust Trajectory Distillation: Hybrid Reweighting Meets Teacher-Inspired Targets","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01920","citing_title":"Pool-Select-Refine for Allocation-Aware Generative Dataset Distillation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31016","citing_title":"An Efficient and Scalable Graph Condensation with Structure-Preserving","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23482","citing_title":"Multimodal Distribution Matching for Vision-Language Dataset Distillation","ref_index":81,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26093","citing_title":"ROAST: Risk-aware Outlier-exposure for Adversarial Selective Training of Anomaly Detectors Against Evasion Attacks","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12649","citing_title":"DIVER:Diving Deeper into Distilled Data via Expressive Semantic Recovery","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11705","citing_title":"CAST: Collapse-Aware multi-Scale Topology Fusion for Multimodal Coreset Selection","ref_index":17,"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":81,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00185","citing_title":"Fair Dataset Distillation via Cross-Group Barycenter Alignment","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12941","citing_title":"Direct Discrepancy Replay: Distribution-Discrepancy Condensation and Manifold-Consistent Replay for Continual Face Forgery Detection","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10666","citing_title":"Omnimodal Dataset Distillation via High-order Proxy Alignment","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ","json":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ.json","graph_json":"https://pith.science/api/pith-number/CMBUID3PCY7KJDJDM6N4OLXWUZ/graph.json","events_json":"https://pith.science/api/pith-number/CMBUID3PCY7KJDJDM6N4OLXWUZ/events.json","paper":"https://pith.science/paper/CMBUID3P"},"agent_actions":{"view_html":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ","download_json":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ.json","view_paper":"https://pith.science/paper/CMBUID3P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2006.05929&json=true","fetch_graph":"https://pith.science/api/pith-number/CMBUID3PCY7KJDJDM6N4OLXWUZ/graph.json","fetch_events":"https://pith.science/api/pith-number/CMBUID3PCY7KJDJDM6N4OLXWUZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ/action/storage_attestation","attest_author":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ/action/author_attestation","sign_citation":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ/action/citation_signature","submit_replication":"https://pith.science/pith/CMBUID3PCY7KJDJDM6N4OLXWUZ/action/replication_record"}},"created_at":"2026-07-05T02:20:55.210644+00:00","updated_at":"2026-07-05T02:20:55.210644+00:00"}