{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IZW7ZTGIZTAAZ2J6MAJTK2C5Q3","short_pith_number":"pith:IZW7ZTGI","schema_version":"1.0","canonical_sha256":"466dfcccc8ccc00ce93e601335685d86ee9e93d2516048629c35837ae0ba4855","source":{"kind":"arxiv","id":"2605.22183","version":1},"attestation_state":"computed","paper":{"title":"Action with Visual Primitives","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Gao Huang, Renping Zhou, Rui Fang, Weilong Guo, Wenda Xu, Yuan He, Yuchen Wang, Yue Meng, Yunfeng Zhang","submitted_at":"2026-05-21T08:52:47Z","abstract_excerpt":"Vision-Language-Action (VLA) models have emerged as a promising paradigm for generalist robotic manipulation. A common design in current architectures maps language instructions and visual observations to actions in a single forward pass. While conceptually simple, this formulation entangles instruction comprehension, spatial scene understanding, and motor control within a single learning objective. As a result, the action expert must implicitly relearn cognitive and perceptual capabilities already present in the pretrained VLM, which can limit both learning efficiency and generalization. We i"},"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":"2605.22183","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-05-21T08:52:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9cfae2ad0cb5360cacc4e26eb2ea546e8bcb67509bd4300811eb3d9027dcdf9c","abstract_canon_sha256":"b2f39331098eff6aaab3e7e8d2f0b31104052613619ce3d75547af74778853d5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:30.718756Z","signature_b64":"9ke8UEE4YNQA/irKS5KY/RmIlmUCmg7KkQwok/h1NcMq/dYJa27IVwowivcnBu7FR7ux2H6DDEMUi1S1iifXBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"466dfcccc8ccc00ce93e601335685d86ee9e93d2516048629c35837ae0ba4855","last_reissued_at":"2026-05-22T01:04:30.717925Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:30.717925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Action with Visual Primitives","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Gao Huang, Renping Zhou, Rui Fang, Weilong Guo, Wenda Xu, Yuan He, Yuchen Wang, Yue Meng, Yunfeng Zhang","submitted_at":"2026-05-21T08:52:47Z","abstract_excerpt":"Vision-Language-Action (VLA) models have emerged as a promising paradigm for generalist robotic manipulation. A common design in current architectures maps language instructions and visual observations to actions in a single forward pass. While conceptually simple, this formulation entangles instruction comprehension, spatial scene understanding, and motor control within a single learning objective. As a result, the action expert must implicitly relearn cognitive and perceptual capabilities already present in the pretrained VLM, which can limit both learning efficiency and generalization. We i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22183","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/2605.22183/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":"2605.22183","created_at":"2026-05-22T01:04:30.718065+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.22183v1","created_at":"2026-05-22T01:04:30.718065+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22183","created_at":"2026-05-22T01:04:30.718065+00:00"},{"alias_kind":"pith_short_12","alias_value":"IZW7ZTGIZTAA","created_at":"2026-05-22T01:04:30.718065+00:00"},{"alias_kind":"pith_short_16","alias_value":"IZW7ZTGIZTAAZ2J6","created_at":"2026-05-22T01:04:30.718065+00:00"},{"alias_kind":"pith_short_8","alias_value":"IZW7ZTGI","created_at":"2026-05-22T01:04:30.718065+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/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3","json":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3.json","graph_json":"https://pith.science/api/pith-number/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/graph.json","events_json":"https://pith.science/api/pith-number/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/events.json","paper":"https://pith.science/paper/IZW7ZTGI"},"agent_actions":{"view_html":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3","download_json":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3.json","view_paper":"https://pith.science/paper/IZW7ZTGI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.22183&json=true","fetch_graph":"https://pith.science/api/pith-number/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/graph.json","fetch_events":"https://pith.science/api/pith-number/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/action/storage_attestation","attest_author":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/action/author_attestation","sign_citation":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/action/citation_signature","submit_replication":"https://pith.science/pith/IZW7ZTGIZTAAZ2J6MAJTK2C5Q3/action/replication_record"}},"created_at":"2026-05-22T01:04:30.718065+00:00","updated_at":"2026-05-22T01:04:30.718065+00:00"}