{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:PPMQZ4CFYIMN6T5IPNJOPWTYLV","short_pith_number":"pith:PPMQZ4CF","schema_version":"1.0","canonical_sha256":"7bd90cf045c218df4fa87b52e7da785d75473e48a6dbb7fc0ff5fad6bb52e57c","source":{"kind":"arxiv","id":"2603.25144","version":2},"attestation_state":"computed","paper":{"title":"FD$^2$: A Dedicated Framework for Fine-Grained Dataset Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Baoli Sun, Dongzhan Zhou, Guang Li, Hongxu Ma, Miki Haseyama, Shijie Wang, Takahiro Ogawa, Zhihui Wang","submitted_at":"2026-03-26T08:03:45Z","abstract_excerpt":"Dataset distillation (DD) compresses a large training set into a small synthetic set, reducing storage and training cost, and has shown strong results on general benchmarks. Decoupled DD further improves efficiency by splitting the pipeline into pretraining, sample distillation, and soft-label generation. However, existing decoupled methods largely rely on coarse class-label supervision and optimize samples within each class in a nearly identical manner. On fine-grained datasets, this often yields distilled samples that (i) retain large intra-class variation with subtle inter-class differences"},"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":"2603.25144","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-03-26T08:03:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c1d8e2c6b7922ac8b9d1e949383f816c950e314301dfab92f1dbe12b05603875","abstract_canon_sha256":"e42b8b39499505976170c88b08c33744cca91bcbd7624959f64e55bed918d625"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:17:37.605270Z","signature_b64":"ljqRXVIrV8I8tx6H3bgkduUtCUVQLfwf4SjyF97SueIBenCT80jmXpIS8zHga5u1aNVPLxrRAxGPqJGfIIt7Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7bd90cf045c218df4fa87b52e7da785d75473e48a6dbb7fc0ff5fad6bb52e57c","last_reissued_at":"2026-06-30T01:17:37.604671Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:17:37.604671Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FD$^2$: A Dedicated Framework for Fine-Grained Dataset Distillation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Baoli Sun, Dongzhan Zhou, Guang Li, Hongxu Ma, Miki Haseyama, Shijie Wang, Takahiro Ogawa, Zhihui Wang","submitted_at":"2026-03-26T08:03:45Z","abstract_excerpt":"Dataset distillation (DD) compresses a large training set into a small synthetic set, reducing storage and training cost, and has shown strong results on general benchmarks. Decoupled DD further improves efficiency by splitting the pipeline into pretraining, sample distillation, and soft-label generation. However, existing decoupled methods largely rely on coarse class-label supervision and optimize samples within each class in a nearly identical manner. On fine-grained datasets, this often yields distilled samples that (i) retain large intra-class variation with subtle inter-class differences"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.25144","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/2603.25144/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":"2603.25144","created_at":"2026-06-30T01:17:37.604748+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.25144v2","created_at":"2026-06-30T01:17:37.604748+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.25144","created_at":"2026-06-30T01:17:37.604748+00:00"},{"alias_kind":"pith_short_12","alias_value":"PPMQZ4CFYIMN","created_at":"2026-06-30T01:17:37.604748+00:00"},{"alias_kind":"pith_short_16","alias_value":"PPMQZ4CFYIMN6T5I","created_at":"2026-06-30T01:17:37.604748+00:00"},{"alias_kind":"pith_short_8","alias_value":"PPMQZ4CF","created_at":"2026-06-30T01:17:37.604748+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.07194","citing_title":"Closed-Form Linear-Probe Dataset Distillation for Pre-trained Vision Models","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV","json":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV.json","graph_json":"https://pith.science/api/pith-number/PPMQZ4CFYIMN6T5IPNJOPWTYLV/graph.json","events_json":"https://pith.science/api/pith-number/PPMQZ4CFYIMN6T5IPNJOPWTYLV/events.json","paper":"https://pith.science/paper/PPMQZ4CF"},"agent_actions":{"view_html":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV","download_json":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV.json","view_paper":"https://pith.science/paper/PPMQZ4CF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.25144&json=true","fetch_graph":"https://pith.science/api/pith-number/PPMQZ4CFYIMN6T5IPNJOPWTYLV/graph.json","fetch_events":"https://pith.science/api/pith-number/PPMQZ4CFYIMN6T5IPNJOPWTYLV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV/action/storage_attestation","attest_author":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV/action/author_attestation","sign_citation":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV/action/citation_signature","submit_replication":"https://pith.science/pith/PPMQZ4CFYIMN6T5IPNJOPWTYLV/action/replication_record"}},"created_at":"2026-06-30T01:17:37.604748+00:00","updated_at":"2026-06-30T01:17:37.604748+00:00"}