{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GS7Q7HFASRY3OS7242KF462MEC","short_pith_number":"pith:GS7Q7HFA","schema_version":"1.0","canonical_sha256":"34bf0f9ca09471b74bfae6945e7b4c20805f63c61e7d49a3dd84d44c3e9a8762","source":{"kind":"arxiv","id":"2503.09871","version":1},"attestation_state":"computed","paper":{"title":"LuciBot: Automated Robot Policy Learning from Generated Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chuang Gan, Chunru Lin, Jiting Cai, Tsun-Hsuan Wang, Xiaowen Qiu, Yian Wang, Zhehuan Chen","submitted_at":"2025-03-12T22:07:36Z","abstract_excerpt":"Automatically generating training supervision for embodied tasks is crucial, as manual designing is tedious and not scalable. While prior works use large language models (LLMs) or vision-language models (VLMs) to generate rewards, these approaches are largely limited to simple tasks with well-defined rewards, such as pick-and-place. This limitation arises because LLMs struggle to interpret complex scenes compressed into text or code due to their restricted input modality, while VLM-based rewards, though better at visual perception, remain limited by their less expressive output modality. To ad"},"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":"2503.09871","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-12T22:07:36Z","cross_cats_sorted":[],"title_canon_sha256":"80ef4b9fd2a086d5342dfd90676bb64e03ce5159214492e63f3a7b97231bdc80","abstract_canon_sha256":"20c6fb322ab4c25d8e656c6af3c02c0057d6a91ec94cbd6fa1ef77cf9cfcd0c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:30:10.457240Z","signature_b64":"T9tZr3V4xxaiX4nwHzMLcFqBHSEF3iulXtBKxTLnF/ND3Upm8ZPNurNPIGR1YXeai+RR9ItNcU3EpmUYkPBbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"34bf0f9ca09471b74bfae6945e7b4c20805f63c61e7d49a3dd84d44c3e9a8762","last_reissued_at":"2026-07-05T10:30:10.456357Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:30:10.456357Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LuciBot: Automated Robot Policy Learning from Generated Videos","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chuang Gan, Chunru Lin, Jiting Cai, Tsun-Hsuan Wang, Xiaowen Qiu, Yian Wang, Zhehuan Chen","submitted_at":"2025-03-12T22:07:36Z","abstract_excerpt":"Automatically generating training supervision for embodied tasks is crucial, as manual designing is tedious and not scalable. While prior works use large language models (LLMs) or vision-language models (VLMs) to generate rewards, these approaches are largely limited to simple tasks with well-defined rewards, such as pick-and-place. This limitation arises because LLMs struggle to interpret complex scenes compressed into text or code due to their restricted input modality, while VLM-based rewards, though better at visual perception, remain limited by their less expressive output modality. To ad"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.09871","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/2503.09871/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":"2503.09871","created_at":"2026-07-05T10:30:10.456433+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.09871v1","created_at":"2026-07-05T10:30:10.456433+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.09871","created_at":"2026-07-05T10:30:10.456433+00:00"},{"alias_kind":"pith_short_12","alias_value":"GS7Q7HFASRY3","created_at":"2026-07-05T10:30:10.456433+00:00"},{"alias_kind":"pith_short_16","alias_value":"GS7Q7HFASRY3OS72","created_at":"2026-07-05T10:30:10.456433+00:00"},{"alias_kind":"pith_short_8","alias_value":"GS7Q7HFA","created_at":"2026-07-05T10:30:10.456433+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24448","citing_title":"Supervise What Survives: Geometry-Guided VLA Adaptation from Synthetic Robot Videos","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2603.28489","citing_title":"Video Generation Models as World Models: Efficient Paradigms, Architectures and Algorithms","ref_index":203,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21741","citing_title":"Hi-WM: Human-in-the-World-Model for Scalable Robot Post-Training","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC","json":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC.json","graph_json":"https://pith.science/api/pith-number/GS7Q7HFASRY3OS7242KF462MEC/graph.json","events_json":"https://pith.science/api/pith-number/GS7Q7HFASRY3OS7242KF462MEC/events.json","paper":"https://pith.science/paper/GS7Q7HFA"},"agent_actions":{"view_html":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC","download_json":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC.json","view_paper":"https://pith.science/paper/GS7Q7HFA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.09871&json=true","fetch_graph":"https://pith.science/api/pith-number/GS7Q7HFASRY3OS7242KF462MEC/graph.json","fetch_events":"https://pith.science/api/pith-number/GS7Q7HFASRY3OS7242KF462MEC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC/action/storage_attestation","attest_author":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC/action/author_attestation","sign_citation":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC/action/citation_signature","submit_replication":"https://pith.science/pith/GS7Q7HFASRY3OS7242KF462MEC/action/replication_record"}},"created_at":"2026-07-05T10:30:10.456433+00:00","updated_at":"2026-07-05T10:30:10.456433+00:00"}