{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DFX4JHD2Z2KRV2TXDGRQXGOWX2","short_pith_number":"pith:DFX4JHD2","schema_version":"1.0","canonical_sha256":"196fc49c7ace951aea7719a30b99d6bea1aaf554d8b5f98f2ce68781a919e653","source":{"kind":"arxiv","id":"2503.10966","version":4},"attestation_state":"computed","paper":{"title":"Is Your Imitation Learning Policy Better than Mine? Policy Comparison with Near-Optimal Stopping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.RO","authors_text":"Anirudha Majumdar, Apurva Badithela, Asher James Hancock, David Snyder, Emma Dixon, Haruki Nishimura, Masha Itkina, Patrick Miller, Rares Andrei Ambrus","submitted_at":"2025-03-14T00:21:48Z","abstract_excerpt":"Imitation learning has enabled robots to perform complex, long-horizon tasks in challenging dexterous manipulation settings. As new methods are developed, they must be rigorously evaluated and compared against corresponding baselines through repeated evaluation trials. However, policy comparison is fundamentally constrained by a small feasible sample size (e.g., 10 or 50) due to significant human effort and limited inference throughput of policies. This paper proposes a novel statistical framework for rigorously comparing two policies in the small sample size regime. Prior work in statistical "},"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":"2503.10966","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-03-14T00:21:48Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"24b5a01ab15db1f68d771be52ac01a9c6389d7ee532fe08ecf7b195277920b44","abstract_canon_sha256":"cf846b7edd57c7df2b5144ff736e042239fdb0ece3df1babd34e389ae1e450c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:17:23.920502Z","signature_b64":"Nj4SRIs5pJi0siOB36+juVvtqcFj3XDUId6Jcq3Pa/5//v7uCNz0v6a1eRd+EfPhOAdfvPqjt6MlIx5FaeAODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"196fc49c7ace951aea7719a30b99d6bea1aaf554d8b5f98f2ce68781a919e653","last_reissued_at":"2026-07-05T11:17:23.919973Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:17:23.919973Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is Your Imitation Learning Policy Better than Mine? Policy Comparison with Near-Optimal Stopping","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.RO","authors_text":"Anirudha Majumdar, Apurva Badithela, Asher James Hancock, David Snyder, Emma Dixon, Haruki Nishimura, Masha Itkina, Patrick Miller, Rares Andrei Ambrus","submitted_at":"2025-03-14T00:21:48Z","abstract_excerpt":"Imitation learning has enabled robots to perform complex, long-horizon tasks in challenging dexterous manipulation settings. As new methods are developed, they must be rigorously evaluated and compared against corresponding baselines through repeated evaluation trials. However, policy comparison is fundamentally constrained by a small feasible sample size (e.g., 10 or 50) due to significant human effort and limited inference throughput of policies. This paper proposes a novel statistical framework for rigorously comparing two policies in the small sample size regime. Prior work in statistical "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.10966","kind":"arxiv","version":4},"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/2503.10966/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":"2503.10966","created_at":"2026-07-05T11:17:23.920044+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.10966v4","created_at":"2026-07-05T11:17:23.920044+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.10966","created_at":"2026-07-05T11:17:23.920044+00:00"},{"alias_kind":"pith_short_12","alias_value":"DFX4JHD2Z2KR","created_at":"2026-07-05T11:17:23.920044+00:00"},{"alias_kind":"pith_short_16","alias_value":"DFX4JHD2Z2KRV2TX","created_at":"2026-07-05T11:17:23.920044+00:00"},{"alias_kind":"pith_short_8","alias_value":"DFX4JHD2","created_at":"2026-07-05T11:17:23.920044+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.24642","citing_title":"Understanding the Impact of Geometric Foundation Models on Vision-Language-Action Models","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29710","citing_title":"PhAIL: A Real-Robot VLA Benchmark and Distributional Methodology","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2507.05331","citing_title":"A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation","ref_index":78,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2","json":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2.json","graph_json":"https://pith.science/api/pith-number/DFX4JHD2Z2KRV2TXDGRQXGOWX2/graph.json","events_json":"https://pith.science/api/pith-number/DFX4JHD2Z2KRV2TXDGRQXGOWX2/events.json","paper":"https://pith.science/paper/DFX4JHD2"},"agent_actions":{"view_html":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2","download_json":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2.json","view_paper":"https://pith.science/paper/DFX4JHD2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.10966&json=true","fetch_graph":"https://pith.science/api/pith-number/DFX4JHD2Z2KRV2TXDGRQXGOWX2/graph.json","fetch_events":"https://pith.science/api/pith-number/DFX4JHD2Z2KRV2TXDGRQXGOWX2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2/action/storage_attestation","attest_author":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2/action/author_attestation","sign_citation":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2/action/citation_signature","submit_replication":"https://pith.science/pith/DFX4JHD2Z2KRV2TXDGRQXGOWX2/action/replication_record"}},"created_at":"2026-07-05T11:17:23.920044+00:00","updated_at":"2026-07-05T11:17:23.920044+00:00"}