{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ON6KY5A265H6SWYHSJUGKRHNXK","short_pith_number":"pith:ON6KY5A2","schema_version":"1.0","canonical_sha256":"737cac741af74fe95b0792686544edba8834b380686a96f203c270eec4c362a3","source":{"kind":"arxiv","id":"2505.22626","version":2},"attestation_state":"computed","paper":{"title":"SCIZOR: A Self-Supervised Approach to Data Curation for Large-Scale Imitation Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Huihan Liu, Linxi Fan, Michael Wan, Rutav Shah, Yuke Zhu, Yuqi Xie, Yu Zhang","submitted_at":"2025-05-28T17:45:05Z","abstract_excerpt":"Imitation learning advances robot capabilities by enabling the acquisition of diverse behaviors from human demonstrations. However, large-scale datasets used for policy training often introduce substantial variability in quality, which can negatively impact performance. As a result, automatically curating datasets by filtering low-quality samples to improve quality becomes essential. Existing robotic curation approaches rely on costly manual annotations and perform curation at a coarse granularity, such as the dataset or trajectory level, failing to account for the quality of individual state-"},"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":"2505.22626","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-05-28T17:45:05Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6113cdd179b1ea45d87316a0e28e08c35b3eedff9c8a1dc3c5265b3233a7e3d5","abstract_canon_sha256":"179e60d74544634589ac470aea9793d12e868433c96d13f22325e3b1f30c805d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:07:07.982384Z","signature_b64":"SP0zvzaAdtL7BqyhMeLDaPXejgI3I3zYxVJ2po7zoYsEqcQ8zVo5IUTmqwgG+z1lFOYaq/02zTfGkAyJD88GBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"737cac741af74fe95b0792686544edba8834b380686a96f203c270eec4c362a3","last_reissued_at":"2026-07-05T12:07:07.981874Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:07:07.981874Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SCIZOR: A Self-Supervised Approach to Data Curation for Large-Scale Imitation Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Huihan Liu, Linxi Fan, Michael Wan, Rutav Shah, Yuke Zhu, Yuqi Xie, Yu Zhang","submitted_at":"2025-05-28T17:45:05Z","abstract_excerpt":"Imitation learning advances robot capabilities by enabling the acquisition of diverse behaviors from human demonstrations. However, large-scale datasets used for policy training often introduce substantial variability in quality, which can negatively impact performance. As a result, automatically curating datasets by filtering low-quality samples to improve quality becomes essential. Existing robotic curation approaches rely on costly manual annotations and perform curation at a coarse granularity, such as the dataset or trajectory level, failing to account for the quality of individual state-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.22626","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/2505.22626/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":"2505.22626","created_at":"2026-07-05T12:07:07.981935+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.22626v2","created_at":"2026-07-05T12:07:07.981935+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.22626","created_at":"2026-07-05T12:07:07.981935+00:00"},{"alias_kind":"pith_short_12","alias_value":"ON6KY5A265H6","created_at":"2026-07-05T12:07:07.981935+00:00"},{"alias_kind":"pith_short_16","alias_value":"ON6KY5A265H6SWYH","created_at":"2026-07-05T12:07:07.981935+00:00"},{"alias_kind":"pith_short_8","alias_value":"ON6KY5A2","created_at":"2026-07-05T12:07:07.981935+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06442","citing_title":"SIEVE: Structure-Aware Data Selection for Imitation Learning with VLA Models","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2606.22142","citing_title":"RoboLineage: Agent-Native Data Lifecycle Governance Across Robot Policy Iterations","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2510.08547","citing_title":"R2RGEN: Real-to-Real 3D Data Generation for Spatially Generalized Manipulation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01544","citing_title":"An Efficient Metric for Data Quality Measurement in Imitation Learning","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK","json":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK.json","graph_json":"https://pith.science/api/pith-number/ON6KY5A265H6SWYHSJUGKRHNXK/graph.json","events_json":"https://pith.science/api/pith-number/ON6KY5A265H6SWYHSJUGKRHNXK/events.json","paper":"https://pith.science/paper/ON6KY5A2"},"agent_actions":{"view_html":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK","download_json":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK.json","view_paper":"https://pith.science/paper/ON6KY5A2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.22626&json=true","fetch_graph":"https://pith.science/api/pith-number/ON6KY5A265H6SWYHSJUGKRHNXK/graph.json","fetch_events":"https://pith.science/api/pith-number/ON6KY5A265H6SWYHSJUGKRHNXK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK/action/storage_attestation","attest_author":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK/action/author_attestation","sign_citation":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK/action/citation_signature","submit_replication":"https://pith.science/pith/ON6KY5A265H6SWYHSJUGKRHNXK/action/replication_record"}},"created_at":"2026-07-05T12:07:07.981935+00:00","updated_at":"2026-07-05T12:07:07.981935+00:00"}