{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MEQH25FG7ODMRD2FW2IL7QT2QL","short_pith_number":"pith:MEQH25FG","schema_version":"1.0","canonical_sha256":"61207d74a6fb86c88f45b690bfc27a82f652af770b5939a2d4ecb65ac4696009","source":{"kind":"arxiv","id":"2606.02742","version":1},"attestation_state":"computed","paper":{"title":"Consistent Yet Wrong: Evidence Insensitivity in Spatial Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"S Divakar Bhat, Toshihiko Yamasaki","submitted_at":"2026-06-01T18:06:08Z","abstract_excerpt":"Spatial reasoning is fundamental to robotics, autonomy, and embodied AI, yet modern vision-language models (VLMs) remain unreliable on metric distance queries. A common assumption is that consistent predictions across viewpoints reflect geometric grounding. We test this assumption and find the opposite: leading VLMs often produce view-invariant and consistent answers even when those answers are incorrect, indicating weak coupling between predictions and viewpoint-specific visual evidence.\n  We introduce \\textbf{ViewDiag}, a controlled multi-view evaluation protocol built from Hypersim, ScanNet"},"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":"2606.02742","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-01T18:06:08Z","cross_cats_sorted":[],"title_canon_sha256":"638a6af46d75e2c861e3f8b1ae6542d6b1281c539acbbc47490b289de16c7f1c","abstract_canon_sha256":"f75d4de469d7ab74082a27980b75329c040fbf667483d7da14e8b757c7523ed3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T00:05:08.576747Z","signature_b64":"dLswrjb7tgzlYozTm1jA2Ip1igcgdK4iBCIWXjRc3FAkpC1I9z1b1RiinRhlsDW7hmVJUrtw9Bi8hPllf20lDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"61207d74a6fb86c88f45b690bfc27a82f652af770b5939a2d4ecb65ac4696009","last_reissued_at":"2026-06-03T00:05:08.576309Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T00:05:08.576309Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Consistent Yet Wrong: Evidence Insensitivity in Spatial Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"S Divakar Bhat, Toshihiko Yamasaki","submitted_at":"2026-06-01T18:06:08Z","abstract_excerpt":"Spatial reasoning is fundamental to robotics, autonomy, and embodied AI, yet modern vision-language models (VLMs) remain unreliable on metric distance queries. A common assumption is that consistent predictions across viewpoints reflect geometric grounding. We test this assumption and find the opposite: leading VLMs often produce view-invariant and consistent answers even when those answers are incorrect, indicating weak coupling between predictions and viewpoint-specific visual evidence.\n  We introduce \\textbf{ViewDiag}, a controlled multi-view evaluation protocol built from Hypersim, ScanNet"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02742","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/2606.02742/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":"2606.02742","created_at":"2026-06-03T00:05:08.576394+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.02742v1","created_at":"2026-06-03T00:05:08.576394+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02742","created_at":"2026-06-03T00:05:08.576394+00:00"},{"alias_kind":"pith_short_12","alias_value":"MEQH25FG7ODM","created_at":"2026-06-03T00:05:08.576394+00:00"},{"alias_kind":"pith_short_16","alias_value":"MEQH25FG7ODMRD2F","created_at":"2026-06-03T00:05:08.576394+00:00"},{"alias_kind":"pith_short_8","alias_value":"MEQH25FG","created_at":"2026-06-03T00:05:08.576394+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/MEQH25FG7ODMRD2FW2IL7QT2QL","json":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL.json","graph_json":"https://pith.science/api/pith-number/MEQH25FG7ODMRD2FW2IL7QT2QL/graph.json","events_json":"https://pith.science/api/pith-number/MEQH25FG7ODMRD2FW2IL7QT2QL/events.json","paper":"https://pith.science/paper/MEQH25FG"},"agent_actions":{"view_html":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL","download_json":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL.json","view_paper":"https://pith.science/paper/MEQH25FG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.02742&json=true","fetch_graph":"https://pith.science/api/pith-number/MEQH25FG7ODMRD2FW2IL7QT2QL/graph.json","fetch_events":"https://pith.science/api/pith-number/MEQH25FG7ODMRD2FW2IL7QT2QL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL/action/storage_attestation","attest_author":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL/action/author_attestation","sign_citation":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL/action/citation_signature","submit_replication":"https://pith.science/pith/MEQH25FG7ODMRD2FW2IL7QT2QL/action/replication_record"}},"created_at":"2026-06-03T00:05:08.576394+00:00","updated_at":"2026-06-03T00:05:08.576394+00:00"}