{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:L4OFJQYFJ42VUVFJYKLCSQRKK3","short_pith_number":"pith:L4OFJQYF","schema_version":"1.0","canonical_sha256":"5f1c54c3054f355a54a9c29629422a56ef112bfcdeb319015f63ea1b596fc963","source":{"kind":"arxiv","id":"2606.20871","version":1},"attestation_state":"computed","paper":{"title":"Geometric Entropy: When Trajectory Diversity Helps and Hurts in Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Pei Zhou, Qian Luo, Ruizhe Liu, Xunzhe Zhou, Yanchao Yang","submitted_at":"2026-06-18T19:02:08Z","abstract_excerpt":"We study how trajectory-shape diversity in demonstrations affects imitation learning (IL) performance across models, tasks, and data scales. We introduce Geometric Entropy (H_G), a task-agnostic metric that quantifies the intrinsic diversity of transit trajectories after normalizing away extrinsic variation, such as goal pose and workspace scale, via target-frame alignment. Across multiple IL architectures and both simulated and real-robot contact-rich manipulation tasks, we observe a consistent inverted-U relationship between success and H_G: increasing geometric diversity improves robustness"},"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":"2606.20871","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-06-18T19:02:08Z","cross_cats_sorted":[],"title_canon_sha256":"792aac711530489bbc78578a32d49d6674506ab88dee5f85a7735130c9397faf","abstract_canon_sha256":"826c1d7bcce5b16d2a4942cb03e5939b1bb41ac6999345df5de145d008db8d1f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:12:01.389510Z","signature_b64":"VhW+uGc+RMf2wXlDIrFM1KJmVrubWOwbnQv9VmcPMgLpBnbLzDPi911fmqtB3999hJcVE8eSmdrS8AL6xEXPAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f1c54c3054f355a54a9c29629422a56ef112bfcdeb319015f63ea1b596fc963","last_reissued_at":"2026-06-23T00:12:01.389135Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:12:01.389135Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Geometric Entropy: When Trajectory Diversity Helps and Hurts in Imitation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Pei Zhou, Qian Luo, Ruizhe Liu, Xunzhe Zhou, Yanchao Yang","submitted_at":"2026-06-18T19:02:08Z","abstract_excerpt":"We study how trajectory-shape diversity in demonstrations affects imitation learning (IL) performance across models, tasks, and data scales. We introduce Geometric Entropy (H_G), a task-agnostic metric that quantifies the intrinsic diversity of transit trajectories after normalizing away extrinsic variation, such as goal pose and workspace scale, via target-frame alignment. Across multiple IL architectures and both simulated and real-robot contact-rich manipulation tasks, we observe a consistent inverted-U relationship between success and H_G: increasing geometric diversity improves robustness"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20871","kind":"arxiv","version":1},"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/2606.20871/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":"2606.20871","created_at":"2026-06-23T00:12:01.389195+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20871v1","created_at":"2026-06-23T00:12:01.389195+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20871","created_at":"2026-06-23T00:12:01.389195+00:00"},{"alias_kind":"pith_short_12","alias_value":"L4OFJQYFJ42V","created_at":"2026-06-23T00:12:01.389195+00:00"},{"alias_kind":"pith_short_16","alias_value":"L4OFJQYFJ42VUVFJ","created_at":"2026-06-23T00:12:01.389195+00:00"},{"alias_kind":"pith_short_8","alias_value":"L4OFJQYF","created_at":"2026-06-23T00:12:01.389195+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3","json":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3.json","graph_json":"https://pith.science/api/pith-number/L4OFJQYFJ42VUVFJYKLCSQRKK3/graph.json","events_json":"https://pith.science/api/pith-number/L4OFJQYFJ42VUVFJYKLCSQRKK3/events.json","paper":"https://pith.science/paper/L4OFJQYF"},"agent_actions":{"view_html":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3","download_json":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3.json","view_paper":"https://pith.science/paper/L4OFJQYF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20871&json=true","fetch_graph":"https://pith.science/api/pith-number/L4OFJQYFJ42VUVFJYKLCSQRKK3/graph.json","fetch_events":"https://pith.science/api/pith-number/L4OFJQYFJ42VUVFJYKLCSQRKK3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3/action/storage_attestation","attest_author":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3/action/author_attestation","sign_citation":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3/action/citation_signature","submit_replication":"https://pith.science/pith/L4OFJQYFJ42VUVFJYKLCSQRKK3/action/replication_record"}},"created_at":"2026-06-23T00:12:01.389195+00:00","updated_at":"2026-06-23T00:12:01.389195+00:00"}