{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YDOZDH7QU4PEG3PWHAHPT4AYGH","short_pith_number":"pith:YDOZDH7Q","schema_version":"1.0","canonical_sha256":"c0dd919ff0a71e436df6380ef9f01831c17050846ae154f4baa98a1f827e1247","source":{"kind":"arxiv","id":"2503.11918","version":1},"attestation_state":"computed","paper":{"title":"Sketch-to-Skill: Bootstrapping Robot Learning with Human Drawn Trajectory Sketches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC"],"primary_cat":"cs.RO","authors_text":"Amisha Bhaskar, Anukriti Singh, Peihong Yu, Pratap Tokekar, Zahiruddin Mahammad","submitted_at":"2025-03-14T23:08:29Z","abstract_excerpt":"Training robotic manipulation policies traditionally requires numerous demonstrations and/or environmental rollouts. While recent Imitation Learning (IL) and Reinforcement Learning (RL) methods have reduced the number of required demonstrations, they still rely on expert knowledge to collect high-quality data, limiting scalability and accessibility. We propose Sketch-to-Skill, a novel framework that leverages human-drawn 2D sketch trajectories to bootstrap and guide RL for robotic manipulation. Our approach extends beyond previous sketch-based methods, which were primarily focused on imitation"},"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":"2503.11918","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-03-14T23:08:29Z","cross_cats_sorted":["cs.AI","cs.HC"],"title_canon_sha256":"fb61f96934e904a58ddf5e3cf71c00ff13cebab6d28164b1a26793960e6f3219","abstract_canon_sha256":"219bed9300f413e50ae9841425b1c1dc717b1cb17b1dff862af3ec8561cab76d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:32:42.542126Z","signature_b64":"Dlr9fFscu0MX05TgXhuRWYRFYfGKVmCsqJZ7jv+5RqK6+A72t1PCviL8LhYTRjBLNZygcJE1I+EZZwW0BzPFCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c0dd919ff0a71e436df6380ef9f01831c17050846ae154f4baa98a1f827e1247","last_reissued_at":"2026-07-05T10:32:42.541652Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:32:42.541652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sketch-to-Skill: Bootstrapping Robot Learning with Human Drawn Trajectory Sketches","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.HC"],"primary_cat":"cs.RO","authors_text":"Amisha Bhaskar, Anukriti Singh, Peihong Yu, Pratap Tokekar, Zahiruddin Mahammad","submitted_at":"2025-03-14T23:08:29Z","abstract_excerpt":"Training robotic manipulation policies traditionally requires numerous demonstrations and/or environmental rollouts. While recent Imitation Learning (IL) and Reinforcement Learning (RL) methods have reduced the number of required demonstrations, they still rely on expert knowledge to collect high-quality data, limiting scalability and accessibility. We propose Sketch-to-Skill, a novel framework that leverages human-drawn 2D sketch trajectories to bootstrap and guide RL for robotic manipulation. Our approach extends beyond previous sketch-based methods, which were primarily focused on imitation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.11918","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/2503.11918/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":"2503.11918","created_at":"2026-07-05T10:32:42.541709+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.11918v1","created_at":"2026-07-05T10:32:42.541709+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.11918","created_at":"2026-07-05T10:32:42.541709+00:00"},{"alias_kind":"pith_short_12","alias_value":"YDOZDH7QU4PE","created_at":"2026-07-05T10:32:42.541709+00:00"},{"alias_kind":"pith_short_16","alias_value":"YDOZDH7QU4PEG3PW","created_at":"2026-07-05T10:32:42.541709+00:00"},{"alias_kind":"pith_short_8","alias_value":"YDOZDH7Q","created_at":"2026-07-05T10:32:42.541709+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2510.01433","citing_title":"AFFORD2ACT: Affordance-Guided Automatic Keypoint Selection for Generalizable and Lightweight Robotic Manipulation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05110","citing_title":"LineRides: Line-Guided Reinforcement Learning for Bicycle Robot Stunts","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05110","citing_title":"LineRides: Line-Guided Reinforcement Learning for Bicycle Robot Stunts","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04811","citing_title":"AnyUser: Translating Sketched User Intent into Domestic Robots","ref_index":63,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH","json":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH.json","graph_json":"https://pith.science/api/pith-number/YDOZDH7QU4PEG3PWHAHPT4AYGH/graph.json","events_json":"https://pith.science/api/pith-number/YDOZDH7QU4PEG3PWHAHPT4AYGH/events.json","paper":"https://pith.science/paper/YDOZDH7Q"},"agent_actions":{"view_html":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH","download_json":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH.json","view_paper":"https://pith.science/paper/YDOZDH7Q","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.11918&json=true","fetch_graph":"https://pith.science/api/pith-number/YDOZDH7QU4PEG3PWHAHPT4AYGH/graph.json","fetch_events":"https://pith.science/api/pith-number/YDOZDH7QU4PEG3PWHAHPT4AYGH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH/action/storage_attestation","attest_author":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH/action/author_attestation","sign_citation":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH/action/citation_signature","submit_replication":"https://pith.science/pith/YDOZDH7QU4PEG3PWHAHPT4AYGH/action/replication_record"}},"created_at":"2026-07-05T10:32:42.541709+00:00","updated_at":"2026-07-05T10:32:42.541709+00:00"}