{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:I4HS3SXI5PTZSVTED5NV4JPZRG","short_pith_number":"pith:I4HS3SXI","schema_version":"1.0","canonical_sha256":"470f2dcae8ebe79956641f5b5e25f989932cf5325ce0958339a75f06e151c1b1","source":{"kind":"arxiv","id":"2410.11571","version":2},"attestation_state":"computed","paper":{"title":"SDS -- See it, Do it, Sorted: Quadruped Skill Synthesis from Single Video Demonstration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Dimitrios Kanoulas, Jeffrey Li, Maria Stamatopoulou","submitted_at":"2024-10-15T13:04:11Z","abstract_excerpt":"Imagine a robot learning locomotion skills from any single video, without labels or reward engineering. We introduce SDS (\"See it. Do it. Sorted.\"), an automated pipeline for skill acquisition from unstructured demonstrations. Using GPT-4o, SDS applies novel prompting techniques, in the form of spatio-temporal grid-based visual encoding ($G_{v}$) and structured input decomposition (SUS). These produce executable reward functions (RF) from the raw input videos. The RFs are used to train PPO policies and are optimized through closed-loop evolution, using training footage and performance metrics "},"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":"2410.11571","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-10-15T13:04:11Z","cross_cats_sorted":[],"title_canon_sha256":"1c54341efa546162682627b2f32c25d7ac267536748241b6956a4dadf8a8c4d7","abstract_canon_sha256":"373753ed03a25e3f91efd0907e7b832c2c409aa2a01a1b53459042d2dd09bcb7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:15.854603Z","signature_b64":"unytVigbQ/59B5vXqOjqZPIJUY+jr7GUJOLQj0NDdVFneMCcX2E3/kzTCaArOIc/8atjI6se4xqpgWRXdqG1Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"470f2dcae8ebe79956641f5b5e25f989932cf5325ce0958339a75f06e151c1b1","last_reissued_at":"2026-07-05T11:56:15.854075Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:15.854075Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SDS -- See it, Do it, Sorted: Quadruped Skill Synthesis from Single Video Demonstration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Dimitrios Kanoulas, Jeffrey Li, Maria Stamatopoulou","submitted_at":"2024-10-15T13:04:11Z","abstract_excerpt":"Imagine a robot learning locomotion skills from any single video, without labels or reward engineering. We introduce SDS (\"See it. Do it. Sorted.\"), an automated pipeline for skill acquisition from unstructured demonstrations. Using GPT-4o, SDS applies novel prompting techniques, in the form of spatio-temporal grid-based visual encoding ($G_{v}$) and structured input decomposition (SUS). These produce executable reward functions (RF) from the raw input videos. The RFs are used to train PPO policies and are optimized through closed-loop evolution, using training footage and performance metrics "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.11571","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/2410.11571/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":"2410.11571","created_at":"2026-07-05T11:56:15.854140+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.11571v2","created_at":"2026-07-05T11:56:15.854140+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.11571","created_at":"2026-07-05T11:56:15.854140+00:00"},{"alias_kind":"pith_short_12","alias_value":"I4HS3SXI5PTZ","created_at":"2026-07-05T11:56:15.854140+00:00"},{"alias_kind":"pith_short_16","alias_value":"I4HS3SXI5PTZSVTE","created_at":"2026-07-05T11:56:15.854140+00:00"},{"alias_kind":"pith_short_8","alias_value":"I4HS3SXI","created_at":"2026-07-05T11:56:15.854140+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.10022","citing_title":"APEX: Action Priors Enable Efficient Exploration for Robust Motion Tracking on Legged Robots","ref_index":15,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG","json":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG.json","graph_json":"https://pith.science/api/pith-number/I4HS3SXI5PTZSVTED5NV4JPZRG/graph.json","events_json":"https://pith.science/api/pith-number/I4HS3SXI5PTZSVTED5NV4JPZRG/events.json","paper":"https://pith.science/paper/I4HS3SXI"},"agent_actions":{"view_html":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG","download_json":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG.json","view_paper":"https://pith.science/paper/I4HS3SXI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.11571&json=true","fetch_graph":"https://pith.science/api/pith-number/I4HS3SXI5PTZSVTED5NV4JPZRG/graph.json","fetch_events":"https://pith.science/api/pith-number/I4HS3SXI5PTZSVTED5NV4JPZRG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG/action/storage_attestation","attest_author":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG/action/author_attestation","sign_citation":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG/action/citation_signature","submit_replication":"https://pith.science/pith/I4HS3SXI5PTZSVTED5NV4JPZRG/action/replication_record"}},"created_at":"2026-07-05T11:56:15.854140+00:00","updated_at":"2026-07-05T11:56:15.854140+00:00"}