{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A5OEAPMVVQLYHARUOWSXY4NABJ","short_pith_number":"pith:A5OEAPMV","schema_version":"1.0","canonical_sha256":"075c403d95ac1783823475a57c71a00a4838ac5e63c05f256d25d3be290510f2","source":{"kind":"arxiv","id":"2503.13162","version":2},"attestation_state":"computed","paper":{"title":"Efficient Imitation under Misspecification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Gokul Swamy, Nicolas Espinosa-Dice, Sanjiban Choudhury, Wen Sun","submitted_at":"2025-03-17T13:35:55Z","abstract_excerpt":"We consider the problem of imitation learning under misspecification: settings where the learner is fundamentally unable to replicate expert behavior everywhere. This is often true in practice due to differences in observation space and action space expressiveness (e.g. perceptual or morphological differences between robots and humans). Given the learner must make some mistakes in the misspecified setting, interaction with the environment is fundamentally required to figure out which mistakes are particularly costly and lead to compounding errors. However, given the computational cost and safe"},"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":"2503.13162","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-17T13:35:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b9cb2b313d8fd92c29d6478cc4e2505667c036c7b79b0d8e1a9f60450aa59dca","abstract_canon_sha256":"b9fd6c56cbb10b9f994ec3716abb137d082336de349a69e4ca675de3c6896c3b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:43:26.844306Z","signature_b64":"L7E0krDGPqsxWnlwzaT0XeyFnt34MW5LgDKegylPfVj0R87s7z/bhyr7MTCt57hXckbIX9p4XWWxJNI7bSIzDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"075c403d95ac1783823475a57c71a00a4838ac5e63c05f256d25d3be290510f2","last_reissued_at":"2026-07-05T10:43:26.843791Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:43:26.843791Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Imitation under Misspecification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Gokul Swamy, Nicolas Espinosa-Dice, Sanjiban Choudhury, Wen Sun","submitted_at":"2025-03-17T13:35:55Z","abstract_excerpt":"We consider the problem of imitation learning under misspecification: settings where the learner is fundamentally unable to replicate expert behavior everywhere. This is often true in practice due to differences in observation space and action space expressiveness (e.g. perceptual or morphological differences between robots and humans). Given the learner must make some mistakes in the misspecified setting, interaction with the environment is fundamentally required to figure out which mistakes are particularly costly and lead to compounding errors. However, given the computational cost and safe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.13162","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/2503.13162/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":"2503.13162","created_at":"2026-07-05T10:43:26.843853+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.13162v2","created_at":"2026-07-05T10:43:26.843853+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.13162","created_at":"2026-07-05T10:43:26.843853+00:00"},{"alias_kind":"pith_short_12","alias_value":"A5OEAPMVVQLY","created_at":"2026-07-05T10:43:26.843853+00:00"},{"alias_kind":"pith_short_16","alias_value":"A5OEAPMVVQLYHARU","created_at":"2026-07-05T10:43:26.843853+00:00"},{"alias_kind":"pith_short_8","alias_value":"A5OEAPMV","created_at":"2026-07-05T10:43:26.843853+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30445","citing_title":"When Does Online Imitation Learning Help in LLM Post-Training? The Role of (Non-)Realizability Beyond Horizon","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09183","citing_title":"Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09183","citing_title":"Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ","json":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ.json","graph_json":"https://pith.science/api/pith-number/A5OEAPMVVQLYHARUOWSXY4NABJ/graph.json","events_json":"https://pith.science/api/pith-number/A5OEAPMVVQLYHARUOWSXY4NABJ/events.json","paper":"https://pith.science/paper/A5OEAPMV"},"agent_actions":{"view_html":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ","download_json":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ.json","view_paper":"https://pith.science/paper/A5OEAPMV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.13162&json=true","fetch_graph":"https://pith.science/api/pith-number/A5OEAPMVVQLYHARUOWSXY4NABJ/graph.json","fetch_events":"https://pith.science/api/pith-number/A5OEAPMVVQLYHARUOWSXY4NABJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ/action/storage_attestation","attest_author":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ/action/author_attestation","sign_citation":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ/action/citation_signature","submit_replication":"https://pith.science/pith/A5OEAPMVVQLYHARUOWSXY4NABJ/action/replication_record"}},"created_at":"2026-07-05T10:43:26.843853+00:00","updated_at":"2026-07-05T10:43:26.843853+00:00"}