{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:CJVIV5OOYENKDYVNC5QO5V3WFI","short_pith_number":"pith:CJVIV5OO","schema_version":"1.0","canonical_sha256":"126a8af5cec11aa1e2ad1760eed7762a1f7074e4ac47013fa6c59919180f1a40","source":{"kind":"arxiv","id":"2106.14440","version":2},"attestation_state":"computed","paper":{"title":"VAT-Mart: Learning Visual Action Trajectory Proposals for Manipulating 3D ARTiculated Objects","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"Hao Dong, Kaichun Mo, Leonidas Guibas, Qingnan Fan, Ruihai Wu, Tianhao Wu, Xuelin Chen, Yan Zhao, Yian Wang, Zizheng Guo","submitted_at":"2021-06-28T07:47:31Z","abstract_excerpt":"Perceiving and manipulating 3D articulated objects (e.g., cabinets, doors) in human environments is an important yet challenging task for future home-assistant robots. The space of 3D articulated objects is exceptionally rich in their myriad semantic categories, diverse shape geometry, and complicated part functionality. Previous works mostly abstract kinematic structure with estimated joint parameters and part poses as the visual representations for manipulating 3D articulated objects. In this paper, we propose object-centric actionable visual priors as a novel perception-interaction handshak"},"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":"2106.14440","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-06-28T07:47:31Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"389bd47432cbc304ef3a2770019691832764f5cd593eafb89320c28760458f26","abstract_canon_sha256":"cce7c8c0400ee79509518ef6b097c6b840bd4bc8f48752c96a1f2e31885c5ff3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:10:36.123819Z","signature_b64":"uj2WqpTyubtKAG6oTpT+rzBgafQPDW530OqUduPsL+3jZq1x1GRpEkVf4bpkemKAGhbs6v4Xu7sn87gR2wO/AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"126a8af5cec11aa1e2ad1760eed7762a1f7074e4ac47013fa6c59919180f1a40","last_reissued_at":"2026-07-05T04:10:36.123373Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:10:36.123373Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VAT-Mart: Learning Visual Action Trajectory Proposals for Manipulating 3D ARTiculated Objects","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"Hao Dong, Kaichun Mo, Leonidas Guibas, Qingnan Fan, Ruihai Wu, Tianhao Wu, Xuelin Chen, Yan Zhao, Yian Wang, Zizheng Guo","submitted_at":"2021-06-28T07:47:31Z","abstract_excerpt":"Perceiving and manipulating 3D articulated objects (e.g., cabinets, doors) in human environments is an important yet challenging task for future home-assistant robots. The space of 3D articulated objects is exceptionally rich in their myriad semantic categories, diverse shape geometry, and complicated part functionality. Previous works mostly abstract kinematic structure with estimated joint parameters and part poses as the visual representations for manipulating 3D articulated objects. In this paper, we propose object-centric actionable visual priors as a novel perception-interaction handshak"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.14440","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/2106.14440/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":"2106.14440","created_at":"2026-07-05T04:10:36.123435+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.14440v2","created_at":"2026-07-05T04:10:36.123435+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.14440","created_at":"2026-07-05T04:10:36.123435+00:00"},{"alias_kind":"pith_short_12","alias_value":"CJVIV5OOYENK","created_at":"2026-07-05T04:10:36.123435+00:00"},{"alias_kind":"pith_short_16","alias_value":"CJVIV5OOYENKDYVN","created_at":"2026-07-05T04:10:36.123435+00:00"},{"alias_kind":"pith_short_8","alias_value":"CJVIV5OO","created_at":"2026-07-05T04:10:36.123435+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27036","citing_title":"RelAfford6D: Relational 6D Affordance Graphs for Constraint-Driven Robotic Manipulation","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17511","citing_title":"MagicSim: A Unified Infrastructure for Executable Embodied Interaction","ref_index":86,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06155","citing_title":"AffordanceVLA: A Vision-Language-Action Model Empowering Action Generation through Affordance-Aware Understanding","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01365","citing_title":"VoxAfford: Multi-Scale Voxel-Token Fusion for Open-Vocabulary 3D Affordance Detection","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2604.22551","citing_title":"QDTraj: Exploration of Diverse Trajectory Primitives for Articulated Objects Robotic Manipulation","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI","json":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI.json","graph_json":"https://pith.science/api/pith-number/CJVIV5OOYENKDYVNC5QO5V3WFI/graph.json","events_json":"https://pith.science/api/pith-number/CJVIV5OOYENKDYVNC5QO5V3WFI/events.json","paper":"https://pith.science/paper/CJVIV5OO"},"agent_actions":{"view_html":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI","download_json":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI.json","view_paper":"https://pith.science/paper/CJVIV5OO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.14440&json=true","fetch_graph":"https://pith.science/api/pith-number/CJVIV5OOYENKDYVNC5QO5V3WFI/graph.json","fetch_events":"https://pith.science/api/pith-number/CJVIV5OOYENKDYVNC5QO5V3WFI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI/action/storage_attestation","attest_author":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI/action/author_attestation","sign_citation":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI/action/citation_signature","submit_replication":"https://pith.science/pith/CJVIV5OOYENKDYVNC5QO5V3WFI/action/replication_record"}},"created_at":"2026-07-05T04:10:36.123435+00:00","updated_at":"2026-07-05T04:10:36.123435+00:00"}