{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TBYNHKOVON6ACW24CEN5M4DERG","short_pith_number":"pith:TBYNHKOV","schema_version":"1.0","canonical_sha256":"9870d3a9d5737c015b5c111bd6706489a95f638798f52588788b94b346ba283b","source":{"kind":"arxiv","id":"2606.13460","version":1},"attestation_state":"computed","paper":{"title":"VISA: VLM-Guided Instance Semantic Auditing for 3D Occupancy World Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dinesh Manocha, Jing Liang, Ruiqi Xian, Xuewei Qi, Yuehan Xian","submitted_at":"2026-06-11T15:15:34Z","abstract_excerpt":"Semantic 3D occupancy provides a voxelized world state for autonomous driving and robot decision making, but object and rare-class errors can affect free-space interpretation, collision checking, and temporal state propagation. We show that a common VLM strategy, aligning 3D voxel or object features with crop-caption embeddings, improves text-space similarity without reliably improving closed-set occupancy mIoU. Motivated by this mismatch, we propose VISA, a training-time semantic auditing approach for existing occupancy world models. VISA queries an offline VLM on a representative crop of eac"},"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":"2606.13460","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-11T15:15:34Z","cross_cats_sorted":[],"title_canon_sha256":"c1fd23de6e983a39f7de3edbc86083442d2c177efedcd5d3331e073172ae9acb","abstract_canon_sha256":"153e80ebe103fcc8480f1c9abc741cc0fb7101c7303cef027c61b981c86b3d82"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-12T01:10:03.729960Z","signature_b64":"+tiHlXn6uFykUyOXvEwbSjDEyp7utj3wD1PavKVVsvq3R6VxWqXwfhy8j5Tl2T/cY/khzcOPCutJ8ZtkoYfRDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9870d3a9d5737c015b5c111bd6706489a95f638798f52588788b94b346ba283b","last_reissued_at":"2026-06-12T01:10:03.728943Z","signature_status":"signed_v1","first_computed_at":"2026-06-12T01:10:03.728943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VISA: VLM-Guided Instance Semantic Auditing for 3D Occupancy World Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dinesh Manocha, Jing Liang, Ruiqi Xian, Xuewei Qi, Yuehan Xian","submitted_at":"2026-06-11T15:15:34Z","abstract_excerpt":"Semantic 3D occupancy provides a voxelized world state for autonomous driving and robot decision making, but object and rare-class errors can affect free-space interpretation, collision checking, and temporal state propagation. We show that a common VLM strategy, aligning 3D voxel or object features with crop-caption embeddings, improves text-space similarity without reliably improving closed-set occupancy mIoU. Motivated by this mismatch, we propose VISA, a training-time semantic auditing approach for existing occupancy world models. VISA queries an offline VLM on a representative crop of eac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.13460","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/2606.13460/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":"2606.13460","created_at":"2026-06-12T01:10:03.729124+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.13460v1","created_at":"2026-06-12T01:10:03.729124+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.13460","created_at":"2026-06-12T01:10:03.729124+00:00"},{"alias_kind":"pith_short_12","alias_value":"TBYNHKOVON6A","created_at":"2026-06-12T01:10:03.729124+00:00"},{"alias_kind":"pith_short_16","alias_value":"TBYNHKOVON6ACW24","created_at":"2026-06-12T01:10:03.729124+00:00"},{"alias_kind":"pith_short_8","alias_value":"TBYNHKOV","created_at":"2026-06-12T01:10:03.729124+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/TBYNHKOVON6ACW24CEN5M4DERG","json":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG.json","graph_json":"https://pith.science/api/pith-number/TBYNHKOVON6ACW24CEN5M4DERG/graph.json","events_json":"https://pith.science/api/pith-number/TBYNHKOVON6ACW24CEN5M4DERG/events.json","paper":"https://pith.science/paper/TBYNHKOV"},"agent_actions":{"view_html":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG","download_json":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG.json","view_paper":"https://pith.science/paper/TBYNHKOV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.13460&json=true","fetch_graph":"https://pith.science/api/pith-number/TBYNHKOVON6ACW24CEN5M4DERG/graph.json","fetch_events":"https://pith.science/api/pith-number/TBYNHKOVON6ACW24CEN5M4DERG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG/action/storage_attestation","attest_author":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG/action/author_attestation","sign_citation":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG/action/citation_signature","submit_replication":"https://pith.science/pith/TBYNHKOVON6ACW24CEN5M4DERG/action/replication_record"}},"created_at":"2026-06-12T01:10:03.729124+00:00","updated_at":"2026-06-12T01:10:03.729124+00:00"}