{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:WAGYI63GJGO2EFVADVVII5IYA4","short_pith_number":"pith:WAGYI63G","schema_version":"1.0","canonical_sha256":"b00d847b66499da216a01d6a847518070651e3be4565253edae2c7a3e68e9cb4","source":{"kind":"arxiv","id":"2402.17768","version":2},"attestation_state":"computed","paper":{"title":"Diffusion Meets DAgger: Supercharging Eye-in-hand Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"Matthew Chang, Pranav Kumar, Saurabh Gupta, Xiaoyu Zhang","submitted_at":"2024-02-27T18:59:18Z","abstract_excerpt":"A common failure mode for policies trained with imitation is compounding execution errors at test time. When the learned policy encounters states that are not present in the expert demonstrations, the policy fails, leading to degenerate behavior. The Dataset Aggregation, or DAgger approach to this problem simply collects more data to cover these failure states. However, in practice, this is often prohibitively expensive. In this work, we propose Diffusion Meets DAgger (DMD), a method to reap the benefits of DAgger without the cost for eye-in-hand imitation learning problems. Instead of collect"},"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":"2402.17768","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-02-27T18:59:18Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"d9833c1409c3b5cf203afa94e4ebfc325251e81cb321376fa3ebf3761cae6f9e","abstract_canon_sha256":"16c2eedf159daff50d7ea5c9b323ccde81311dca31f30740ac951d575441bba5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:46.763060Z","signature_b64":"B5gJ4uRL/GzdGjkJ6J2S/Q+PZEcIt3spWzrICdxHDrPT5wTa6GIJ6tVKTutqC/bfLbaRh9dw33yw9muqteANDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b00d847b66499da216a01d6a847518070651e3be4565253edae2c7a3e68e9cb4","last_reissued_at":"2026-07-05T08:27:46.762553Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:46.762553Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diffusion Meets DAgger: Supercharging Eye-in-hand Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"Matthew Chang, Pranav Kumar, Saurabh Gupta, Xiaoyu Zhang","submitted_at":"2024-02-27T18:59:18Z","abstract_excerpt":"A common failure mode for policies trained with imitation is compounding execution errors at test time. When the learned policy encounters states that are not present in the expert demonstrations, the policy fails, leading to degenerate behavior. The Dataset Aggregation, or DAgger approach to this problem simply collects more data to cover these failure states. However, in practice, this is often prohibitively expensive. In this work, we propose Diffusion Meets DAgger (DMD), a method to reap the benefits of DAgger without the cost for eye-in-hand imitation learning problems. Instead of collect"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.17768","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/2402.17768/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":"2402.17768","created_at":"2026-07-05T08:27:46.762618+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.17768v2","created_at":"2026-07-05T08:27:46.762618+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.17768","created_at":"2026-07-05T08:27:46.762618+00:00"},{"alias_kind":"pith_short_12","alias_value":"WAGYI63GJGO2","created_at":"2026-07-05T08:27:46.762618+00:00"},{"alias_kind":"pith_short_16","alias_value":"WAGYI63GJGO2EFVA","created_at":"2026-07-05T08:27:46.762618+00:00"},{"alias_kind":"pith_short_8","alias_value":"WAGYI63G","created_at":"2026-07-05T08:27:46.762618+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19586","citing_title":"One Demo is Worth a Thousand Trajectories: Action-View Augmentation for Visuomotor Policies","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27114","citing_title":"VR-DAgger: Immersive VR for Dexterous Data Collection and Uncertainty-Guided On-Policy Correction","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11762","citing_title":"NavOL: Navigation Policy with Online Imitation Learning","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10809","citing_title":"WARPED: Wrist-Aligned Rendering for Robot Policy Learning from Egocentric Human Demonstrations","ref_index":128,"is_internal_anchor":false},{"citing_arxiv_id":"2604.11351","citing_title":"WM-DAgger: Enabling Efficient Data Aggregation for Imitation Learning with World Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15938","citing_title":"VADF: Vision-Adaptive Diffusion Policy Framework for Efficient Robotic Manipulation","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4","json":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4.json","graph_json":"https://pith.science/api/pith-number/WAGYI63GJGO2EFVADVVII5IYA4/graph.json","events_json":"https://pith.science/api/pith-number/WAGYI63GJGO2EFVADVVII5IYA4/events.json","paper":"https://pith.science/paper/WAGYI63G"},"agent_actions":{"view_html":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4","download_json":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4.json","view_paper":"https://pith.science/paper/WAGYI63G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.17768&json=true","fetch_graph":"https://pith.science/api/pith-number/WAGYI63GJGO2EFVADVVII5IYA4/graph.json","fetch_events":"https://pith.science/api/pith-number/WAGYI63GJGO2EFVADVVII5IYA4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4/action/storage_attestation","attest_author":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4/action/author_attestation","sign_citation":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4/action/citation_signature","submit_replication":"https://pith.science/pith/WAGYI63GJGO2EFVADVVII5IYA4/action/replication_record"}},"created_at":"2026-07-05T08:27:46.762618+00:00","updated_at":"2026-07-05T08:27:46.762618+00:00"}