{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:7BZCR7KEOEOHVZG2V54UPM23EL","short_pith_number":"pith:7BZCR7KE","schema_version":"1.0","canonical_sha256":"f87228fd44711c7ae4daaf7947b35b22e456830225e50145e559343be965aa6c","source":{"kind":"arxiv","id":"2606.17408","version":1},"attestation_state":"computed","paper":{"title":"Where Should Action Generation Begin? A Learnable Source Prior for Generative Robot Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"He-Yang Xu, Meipo Dai, Qi Dou, Qiyuan Zhuang, Xiu-Shen Wei, Yijun Wang, Ying-Jie Shuai","submitted_at":"2026-06-16T01:43:56Z","abstract_excerpt":"Generative robot policies typically begin action generation from an observation-independent standard Gaussian distribution, leaving the choice of source distribution underexplored. This work asks a simple question: where should action generation begin? We propose LeaP, a Learnable source Prior that replaces the standard Gaussian with a proprioception-conditioned diagonal Gaussian over action chunks. Parameterized by a lightweight MLP, LeaP jointly predicts the mean and state-adaptive variance of the source distribution, while keeping the downstream generator architecture and inference solver u"},"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":"2606.17408","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-06-16T01:43:56Z","cross_cats_sorted":["cs.CV","cs.LG"],"title_canon_sha256":"ba0ec2c3fdd884628e53294b2cb380fd98445589cb2e101eefb8485324f23834","abstract_canon_sha256":"3b0dfba2da5a62e18566f315d16767b9f6bdb85a00a0270c5ac9cea923d3542e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:10:12.111149Z","signature_b64":"LeTaWXVf4VDk4T+sATCmMAyd5nkTriiZ5pNaX6NH3aqa3UFvO3cHg0jFgVhxDYQgFXjTHtLwZ6J35QlamuFADA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f87228fd44711c7ae4daaf7947b35b22e456830225e50145e559343be965aa6c","last_reissued_at":"2026-06-19T16:10:12.110811Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:10:12.110811Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Where Should Action Generation Begin? A Learnable Source Prior for Generative Robot Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","cs.LG"],"primary_cat":"cs.RO","authors_text":"He-Yang Xu, Meipo Dai, Qi Dou, Qiyuan Zhuang, Xiu-Shen Wei, Yijun Wang, Ying-Jie Shuai","submitted_at":"2026-06-16T01:43:56Z","abstract_excerpt":"Generative robot policies typically begin action generation from an observation-independent standard Gaussian distribution, leaving the choice of source distribution underexplored. This work asks a simple question: where should action generation begin? We propose LeaP, a Learnable source Prior that replaces the standard Gaussian with a proprioception-conditioned diagonal Gaussian over action chunks. Parameterized by a lightweight MLP, LeaP jointly predicts the mean and state-adaptive variance of the source distribution, while keeping the downstream generator architecture and inference solver u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.17408","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/2606.17408/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":"2606.17408","created_at":"2026-06-19T16:10:12.110871+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.17408v1","created_at":"2026-06-19T16:10:12.110871+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.17408","created_at":"2026-06-19T16:10:12.110871+00:00"},{"alias_kind":"pith_short_12","alias_value":"7BZCR7KEOEOH","created_at":"2026-06-19T16:10:12.110871+00:00"},{"alias_kind":"pith_short_16","alias_value":"7BZCR7KEOEOHVZG2","created_at":"2026-06-19T16:10:12.110871+00:00"},{"alias_kind":"pith_short_8","alias_value":"7BZCR7KE","created_at":"2026-06-19T16:10:12.110871+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/7BZCR7KEOEOHVZG2V54UPM23EL","json":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL.json","graph_json":"https://pith.science/api/pith-number/7BZCR7KEOEOHVZG2V54UPM23EL/graph.json","events_json":"https://pith.science/api/pith-number/7BZCR7KEOEOHVZG2V54UPM23EL/events.json","paper":"https://pith.science/paper/7BZCR7KE"},"agent_actions":{"view_html":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL","download_json":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL.json","view_paper":"https://pith.science/paper/7BZCR7KE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.17408&json=true","fetch_graph":"https://pith.science/api/pith-number/7BZCR7KEOEOHVZG2V54UPM23EL/graph.json","fetch_events":"https://pith.science/api/pith-number/7BZCR7KEOEOHVZG2V54UPM23EL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL/action/storage_attestation","attest_author":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL/action/author_attestation","sign_citation":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL/action/citation_signature","submit_replication":"https://pith.science/pith/7BZCR7KEOEOHVZG2V54UPM23EL/action/replication_record"}},"created_at":"2026-06-19T16:10:12.110871+00:00","updated_at":"2026-06-19T16:10:12.110871+00:00"}