{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:F2IM7JQH2CMVRZMOUM7CVC2V4T","short_pith_number":"pith:F2IM7JQH","schema_version":"1.0","canonical_sha256":"2e90cfa607d09958e58ea33e2a8b55e4d6a61d4ffe7ffe08826a11bafe8e15b4","source":{"kind":"arxiv","id":"2509.02175","version":2},"attestation_state":"computed","paper":{"title":"Understanding Space Is Rocket Science -- Only Top Reasoning Models Can Solve Spatial Understanding Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anthony Ventresque, Ellen Rushe, Mayug Maniparambil, Nils Hoehing, Noel E. O'Connor","submitted_at":"2025-09-02T10:32:58Z","abstract_excerpt":"We propose RocketScience, an open-source contrastive VLM benchmark that tests for spatial relation understanding. It is comprised of entirely new real-world image-text pairs covering mostly relative spatial understanding and the order of objects. The benchmark is designed to be very easy for humans and hard for the current generation of VLMs, and this is empirically verified. Our results show a striking lack of spatial relation understanding in open source and frontier commercial VLMs and a surprisingly high performance of reasoning models. Additionally, we perform a disentanglement analysis t"},"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":"2509.02175","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-09-02T10:32:58Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"02938559bc9c06033fa37e3412ce63597cabd0e58f4eec1af551098713c4b42f","abstract_canon_sha256":"0115a9ad642bdc7d993a03bb3eb8e744823011f70d0c277df4e055cb52b4d0a1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:04:41.050433Z","signature_b64":"3v6WQupeF8b/agd6gT9K1L8bfurtzTVVfVwM16COnWi0cAdXGa4xua3Ei69c6FlO7CtYw7XBOI/0gVKNTyqKAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2e90cfa607d09958e58ea33e2a8b55e4d6a61d4ffe7ffe08826a11bafe8e15b4","last_reissued_at":"2026-07-05T12:04:41.049849Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:04:41.049849Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding Space Is Rocket Science -- Only Top Reasoning Models Can Solve Spatial Understanding Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anthony Ventresque, Ellen Rushe, Mayug Maniparambil, Nils Hoehing, Noel E. O'Connor","submitted_at":"2025-09-02T10:32:58Z","abstract_excerpt":"We propose RocketScience, an open-source contrastive VLM benchmark that tests for spatial relation understanding. It is comprised of entirely new real-world image-text pairs covering mostly relative spatial understanding and the order of objects. The benchmark is designed to be very easy for humans and hard for the current generation of VLMs, and this is empirically verified. Our results show a striking lack of spatial relation understanding in open source and frontier commercial VLMs and a surprisingly high performance of reasoning models. Additionally, we perform a disentanglement analysis t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.02175","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/2509.02175/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":"2509.02175","created_at":"2026-07-05T12:04:41.049915+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.02175v2","created_at":"2026-07-05T12:04:41.049915+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.02175","created_at":"2026-07-05T12:04:41.049915+00:00"},{"alias_kind":"pith_short_12","alias_value":"F2IM7JQH2CMV","created_at":"2026-07-05T12:04:41.049915+00:00"},{"alias_kind":"pith_short_16","alias_value":"F2IM7JQH2CMVRZMO","created_at":"2026-07-05T12:04:41.049915+00:00"},{"alias_kind":"pith_short_8","alias_value":"F2IM7JQH","created_at":"2026-07-05T12:04:41.049915+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/F2IM7JQH2CMVRZMOUM7CVC2V4T","json":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T.json","graph_json":"https://pith.science/api/pith-number/F2IM7JQH2CMVRZMOUM7CVC2V4T/graph.json","events_json":"https://pith.science/api/pith-number/F2IM7JQH2CMVRZMOUM7CVC2V4T/events.json","paper":"https://pith.science/paper/F2IM7JQH"},"agent_actions":{"view_html":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T","download_json":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T.json","view_paper":"https://pith.science/paper/F2IM7JQH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.02175&json=true","fetch_graph":"https://pith.science/api/pith-number/F2IM7JQH2CMVRZMOUM7CVC2V4T/graph.json","fetch_events":"https://pith.science/api/pith-number/F2IM7JQH2CMVRZMOUM7CVC2V4T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T/action/storage_attestation","attest_author":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T/action/author_attestation","sign_citation":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T/action/citation_signature","submit_replication":"https://pith.science/pith/F2IM7JQH2CMVRZMOUM7CVC2V4T/action/replication_record"}},"created_at":"2026-07-05T12:04:41.049915+00:00","updated_at":"2026-07-05T12:04:41.049915+00:00"}