{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:OPU5XVSM6NYIFHPTZH7P2C5HGY","short_pith_number":"pith:OPU5XVSM","schema_version":"1.0","canonical_sha256":"73e9dbd64cf370829df3c9fefd0ba7361186310c904403a2faaa9e5b7afe94cd","source":{"kind":"arxiv","id":"2603.22435","version":2},"attestation_state":"computed","paper":{"title":"CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Dantong Niu, Ethan Kou, Fei-Fei Li, Guanya Shi, Guanzhi Wang, Haoru Xue, Huang Huang, Jiajun Wu, Justin Yu, Karim El-Refai, Ken Goldberg, Letian Fu, Linxi \"Jim\" Fan, Shankar Sastry, Wenli Xiao, Yuke Zhu","submitted_at":"2026-03-23T18:08:10Z","abstract_excerpt":"\"Code-as-Policy\" considers how executable code can complement data-intensive Vision-Language-Action (VLA) methods, yet their effectiveness as autonomous controllers for embodied manipulation remains underexplored. We present CaP-X, an open-access framework for systematically studying Code-as-Policy agents in robot manipulation. At its core is CaP-Gym, an interactive environment in which agents control robots by synthesizing and executing programs that compose perception and control primitives. Building on this foundation, CaP-Bench evaluates frontier language and vision-language models across "},"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":"2603.22435","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-03-23T18:08:10Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"083db97bce9952b492fb33cf110bbf78e7f0967c01f902c64962e5c4dacfc1ad","abstract_canon_sha256":"cae4f85c645d16ac343e9ee25eede2f136d35ce3138352894e2417b1a4ff5996"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-03T01:17:54.910679Z","signature_b64":"q4IT9lyYoC1AZawRCy2eO/wZJMxkbf0uiLlz7JrOo1M+tfx9qX2LnkzsdPeEFhFuEZD41HXoQYud9un/fp3ICQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"73e9dbd64cf370829df3c9fefd0ba7361186310c904403a2faaa9e5b7afe94cd","last_reissued_at":"2026-07-03T01:17:54.910170Z","signature_status":"signed_v1","first_computed_at":"2026-07-03T01:17:54.910170Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CaP-X: A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.RO","authors_text":"Dantong Niu, Ethan Kou, Fei-Fei Li, Guanya Shi, Guanzhi Wang, Haoru Xue, Huang Huang, Jiajun Wu, Justin Yu, Karim El-Refai, Ken Goldberg, Letian Fu, Linxi \"Jim\" Fan, Shankar Sastry, Wenli Xiao, Yuke Zhu","submitted_at":"2026-03-23T18:08:10Z","abstract_excerpt":"\"Code-as-Policy\" considers how executable code can complement data-intensive Vision-Language-Action (VLA) methods, yet their effectiveness as autonomous controllers for embodied manipulation remains underexplored. We present CaP-X, an open-access framework for systematically studying Code-as-Policy agents in robot manipulation. At its core is CaP-Gym, an interactive environment in which agents control robots by synthesizing and executing programs that compose perception and control primitives. Building on this foundation, CaP-Bench evaluates frontier language and vision-language models across "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.22435","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/2603.22435/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":"2603.22435","created_at":"2026-07-03T01:17:54.910230+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.22435v2","created_at":"2026-07-03T01:17:54.910230+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.22435","created_at":"2026-07-03T01:17:54.910230+00:00"},{"alias_kind":"pith_short_12","alias_value":"OPU5XVSM6NYI","created_at":"2026-07-03T01:17:54.910230+00:00"},{"alias_kind":"pith_short_16","alias_value":"OPU5XVSM6NYIFHPT","created_at":"2026-07-03T01:17:54.910230+00:00"},{"alias_kind":"pith_short_8","alias_value":"OPU5XVSM","created_at":"2026-07-03T01:17:54.910230+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":8,"sample":[{"citing_arxiv_id":"2606.20683","citing_title":"From Question Answering to Task Completion: A Survey on Agent System and Harness Design","ref_index":179,"is_internal_anchor":true},{"citing_arxiv_id":"2606.13675","citing_title":"Improving Robotic Generalist Policies via Flow Reversal Steering","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2606.12366","citing_title":"APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies","ref_index":63,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07723","citing_title":"VoLo: A Physical Orchestrator for Open-Vocabulary Long-Horizon Manipulation","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2606.00998","citing_title":"GraspGen-X: Cross-Embodiment 6-DOF Diffusion-based Grasping","ref_index":20,"is_internal_anchor":true},{"citing_arxiv_id":"2606.30613","citing_title":"Sequential Planning via Anchored Robotic Keypoints","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09218","citing_title":"Flame3D: Zero-shot Compositional Reasoning of 3D Scenes with Agentic Language Models","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10096","citing_title":"ABot-Claw: A Foundation for Persistent, Cooperative, and Self-Evolving Robotic Agents","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY","json":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY.json","graph_json":"https://pith.science/api/pith-number/OPU5XVSM6NYIFHPTZH7P2C5HGY/graph.json","events_json":"https://pith.science/api/pith-number/OPU5XVSM6NYIFHPTZH7P2C5HGY/events.json","paper":"https://pith.science/paper/OPU5XVSM"},"agent_actions":{"view_html":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY","download_json":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY.json","view_paper":"https://pith.science/paper/OPU5XVSM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.22435&json=true","fetch_graph":"https://pith.science/api/pith-number/OPU5XVSM6NYIFHPTZH7P2C5HGY/graph.json","fetch_events":"https://pith.science/api/pith-number/OPU5XVSM6NYIFHPTZH7P2C5HGY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY/action/storage_attestation","attest_author":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY/action/author_attestation","sign_citation":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY/action/citation_signature","submit_replication":"https://pith.science/pith/OPU5XVSM6NYIFHPTZH7P2C5HGY/action/replication_record"}},"created_at":"2026-07-03T01:17:54.910230+00:00","updated_at":"2026-07-03T01:17:54.910230+00:00"}